Datasets:

---

ftopal

/

huggingface-datasets

like 0

[ "region:us" ]

[]

[]

[ "# Dataset Card for Evaluation run of SCE/Mistral-7B-summ-ia3-tuned\n\n\n\nDataset automatically created during the evaluation run of model SCE/Mistral-7B-summ-ia3-tuned on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-29T08:13:56.817298(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "TAGS\n#region-us \n", "# Dataset Card for Evaluation run of SCE/Mistral-7B-summ-ia3-tuned\n\n\n\nDataset automatically created during the evaluation run of model SCE/Mistral-7B-summ-ia3-tuned on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-29T08:13:56.817298(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "task_categories:text-classification", "size_categories:100M<n<1B", "region:us" ]

[]

[]

[ "# URL News Dataset Repository\n\nURL", "## Introduction\n\nWelcome to the URL News Dataset Repository! This repository is created by URL and is dedicated to providing free datasets of publicly available news articles. We release new datasets weekly, each containing around 1,000 news articles focused on various themes, topics, or metadata characteristics like sentiment analysis, and top IPTC categories such as finance, sports, and politics.", "## Dataset Overview\n\n- Weekly Releases: New dataset available every week.\n- Thematic Focus: Datasets based on specific themes, topics, or metadata.\n- Rich Metadata: Includes sentiment analysis, categories, publication dates.\n- Diverse Sources: Articles from a wide range of news websites.", "## Usage\n\nThe datasets are free for academic, research, and journalistic purposes:\n\n- Data Analysis: For statistical analyses, trend identification, and pattern recognition.\n- Machine Learning: Suitable for training NLP models, sentiment analysis, etc.\n- Journalistic Research: Helps journalists in data-driven storytelling.", "## Accessing the Datasets\n\n- Browse the repository.\n- Find a dataset that suits your needs.\n- Download the dataset with its detailed description and metadata file.", "## Contribution\n\nContributions are welcome! If you have suggestions or want to contribute, please open an issue or a pull request.", "## Support\n\nFor questions or support, raise an issue in the repository.\n\n\n---" ]

[ "TAGS\n#task_categories-text-classification #size_categories-100M<n<1B #region-us \n", "# URL News Dataset Repository\n\nURL", "## Introduction\n\nWelcome to the URL News Dataset Repository! This repository is created by URL and is dedicated to providing free datasets of publicly available news articles. We release new datasets weekly, each containing around 1,000 news articles focused on various themes, topics, or metadata characteristics like sentiment analysis, and top IPTC categories such as finance, sports, and politics.", "## Dataset Overview\n\n- Weekly Releases: New dataset available every week.\n- Thematic Focus: Datasets based on specific themes, topics, or metadata.\n- Rich Metadata: Includes sentiment analysis, categories, publication dates.\n- Diverse Sources: Articles from a wide range of news websites.", "## Usage\n\nThe datasets are free for academic, research, and journalistic purposes:\n\n- Data Analysis: For statistical analyses, trend identification, and pattern recognition.\n- Machine Learning: Suitable for training NLP models, sentiment analysis, etc.\n- Journalistic Research: Helps journalists in data-driven storytelling.", "## Accessing the Datasets\n\n- Browse the repository.\n- Find a dataset that suits your needs.\n- Download the dataset with its detailed description and metadata file.", "## Contribution\n\nContributions are welcome! If you have suggestions or want to contribute, please open an issue or a pull request.", "## Support\n\nFor questions or support, raise an issue in the repository.\n\n\n---" ]

[ "task_categories:other", "task_categories:text-to-speech", "task_categories:audio-to-audio", "size_categories:1K<n<10K", "language:ja", "license:gpl-3.0", "pjsk", "草薙寧々", "草薙宁宁", "プロセカ", "prsk", "宁宁", "region:us" ]

[]

[ "ja" ]

[ "# 简介\n\n* 草薙寧々 干声带标签数据集\n\n本数据集只收集了游戏内的一部分，并不是全部的宁宁干声语音，其中 nene_org.txt 是标签文件\n\npjsk 全部角色干声带标签数据集的话可以加QQ群：691795641，群公告里有网盘地址", "# 参考\n\n* 声源归属：草薙寧々(CV:Machico)-「プロジェクトセカイ カラフルステージ！ feat. 初音ミク」", "# TODO\n\n* （长期）补全宁宁语音，规范数据集格式。" ]

[ "TAGS\n#task_categories-other #task_categories-text-to-speech #task_categories-audio-to-audio #size_categories-1K<n<10K #language-Japanese #license-gpl-3.0 #pjsk #草薙寧々 #草薙宁宁 #プロセカ #prsk #宁宁 #region-us \n", "# 简介\n\n* 草薙寧々 干声带标签数据集\n\n本数据集只收集了游戏内的一部分，并不是全部的宁宁干声语音，其中 nene_org.txt 是标签文件\n\npjsk 全部角色干声带标签数据集的话可以加QQ群：691795641，群公告里有网盘地址", "# 参考\n\n* 声源归属：草薙寧々(CV:Machico)-「プロジェクトセカイ カラフルステージ！ feat. 初音ミク」", "# TODO\n\n* （长期）补全宁宁语音，规范数据集格式。" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for Evaluation run of SCE/Mistral-7B-summ-ia3-pruned10\n\n\n\nDataset automatically created during the evaluation run of model SCE/Mistral-7B-summ-ia3-pruned10 on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-29T08:19:37.916430(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "TAGS\n#region-us \n", "# Dataset Card for Evaluation run of SCE/Mistral-7B-summ-ia3-pruned10\n\n\n\nDataset automatically created during the evaluation run of model SCE/Mistral-7B-summ-ia3-pruned10 on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-29T08:19:37.916430(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for Evaluation run of SCE/Mistral-7B-summ-ia3-pruned20\n\n\n\nDataset automatically created during the evaluation run of model SCE/Mistral-7B-summ-ia3-pruned20 on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-29T08:25:43.679587(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "TAGS\n#region-us \n", "# Dataset Card for Evaluation run of SCE/Mistral-7B-summ-ia3-pruned20\n\n\n\nDataset automatically created during the evaluation run of model SCE/Mistral-7B-summ-ia3-pruned20 on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-29T08:25:43.679587(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for Evaluation run of SCE/Mistral-7B-summ-lora-tuned\n\n\n\nDataset automatically created during the evaluation run of model SCE/Mistral-7B-summ-lora-tuned on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-29T08:32:02.958349(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "TAGS\n#region-us \n", "# Dataset Card for Evaluation run of SCE/Mistral-7B-summ-lora-tuned\n\n\n\nDataset automatically created during the evaluation run of model SCE/Mistral-7B-summ-lora-tuned on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-29T08:32:02.958349(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "multilinguality:monolingual", "language:fr", "license:other", "region:us" ]

[]

[ "fr" ]

[ "# Dataset Card for ESSAI", "## Dataset Description\n\n- Homepage: URL\n- Pubmed: False\n- Public: False\n- Tasks: TXTCLASS\n\n\nWe manually annotated two corpora from the biomedical field. The ESSAI corpus contains clinical trial protocols in French. They were mainly obtained from the National Cancer Institute The typical protocol consists of two parts: the summary of the trial, which indicates the purpose of the trial and the methods applied; and a detailed description of the trial with the inclusion and exclusion criteria. The CAS corpus contains clinical cases published in scientific literature and training material. They are published in different journals from French-speaking countries (France, Belgium, Switzerland, Canada, African countries, tropical countries) and are related to various medical specialties (cardiology, urology, oncology, obstetrics, pulmonology, gastro-enterology). The purpose of clinical cases is to describe clinical situations of patients. Hence, their content is close to the content of clinical narratives (description of diagnoses, treatments or procedures, evolution, family history, expected audience, etc.). In clinical cases, the negation is frequently used for describing the patient signs, symptoms, and diagnosis. Speculation is present as well but less frequently.\n\nThis version only contain the annotated ESSAI corpus" ]

[ "TAGS\n#multilinguality-monolingual #language-French #license-other #region-us \n", "# Dataset Card for ESSAI", "## Dataset Description\n\n- Homepage: URL\n- Pubmed: False\n- Public: False\n- Tasks: TXTCLASS\n\n\nWe manually annotated two corpora from the biomedical field. The ESSAI corpus contains clinical trial protocols in French. They were mainly obtained from the National Cancer Institute The typical protocol consists of two parts: the summary of the trial, which indicates the purpose of the trial and the methods applied; and a detailed description of the trial with the inclusion and exclusion criteria. The CAS corpus contains clinical cases published in scientific literature and training material. They are published in different journals from French-speaking countries (France, Belgium, Switzerland, Canada, African countries, tropical countries) and are related to various medical specialties (cardiology, urology, oncology, obstetrics, pulmonology, gastro-enterology). The purpose of clinical cases is to describe clinical situations of patients. Hence, their content is close to the content of clinical narratives (description of diagnoses, treatments or procedures, evolution, family history, expected audience, etc.). In clinical cases, the negation is frequently used for describing the patient signs, symptoms, and diagnosis. Speculation is present as well but less frequently.\n\nThis version only contain the annotated ESSAI corpus" ]

[ "task_categories:text-generation", "task_categories:conversational", "language:en", "license:mit", "chatml", "trl", "conversational", "region:us" ]

[]

[ "en" ]

[]

[ "TAGS\n#task_categories-text-generation #task_categories-conversational #language-English #license-mit #chatml #trl #conversational #region-us \n" ]

[ "multilinguality:multilingual", "language:en", "language:zh", "language:ja", "license:cc-by-4.0", "region:us" ]

[]

[ "en", "zh", "ja" ]

[ "# Dataset Card for NTCIR-13 MedWeb", "## Dataset Description\n\n- Homepage: URL\n- Pubmed: False\n- Public: False\n- Tasks: TRANSL,TXTCLASS\n\n\nNTCIR-13 MedWeb (Medical Natural Language Processing for Web Document) task requires\nto perform a multi-label classification that labels for eight diseases/symptoms must\nbe assigned to each tweet. Given pseudo-tweets, the output are Positive:p or Negative:n\nlabels for eight diseases/symptoms. The achievements of this task can almost be\ndirectly applied to a fundamental engine for actual applications.\n\nThis task provides pseudo-Twitter messages in a cross-language and multi-label corpus,\ncovering three languages (Japanese, English, and Chinese), and annotated with eight\nlabels such as influenza, diarrhea/stomachache, hay fever, cough/sore throat, headache,\nfever, runny nose, and cold.\n\nFor more information, see:\nURL\n\nAs this dataset also provides a parallel corpus of pseudo-tweets for english,\njapanese and chinese it can also be used to train translation models between\nthese three languages." ]

[ "TAGS\n#multilinguality-multilingual #language-English #language-Chinese #language-Japanese #license-cc-by-4.0 #region-us \n", "# Dataset Card for NTCIR-13 MedWeb", "## Dataset Description\n\n- Homepage: URL\n- Pubmed: False\n- Public: False\n- Tasks: TRANSL,TXTCLASS\n\n\nNTCIR-13 MedWeb (Medical Natural Language Processing for Web Document) task requires\nto perform a multi-label classification that labels for eight diseases/symptoms must\nbe assigned to each tweet. Given pseudo-tweets, the output are Positive:p or Negative:n\nlabels for eight diseases/symptoms. The achievements of this task can almost be\ndirectly applied to a fundamental engine for actual applications.\n\nThis task provides pseudo-Twitter messages in a cross-language and multi-label corpus,\ncovering three languages (Japanese, English, and Chinese), and annotated with eight\nlabels such as influenza, diarrhea/stomachache, hay fever, cough/sore throat, headache,\nfever, runny nose, and cold.\n\nFor more information, see:\nURL\n\nAs this dataset also provides a parallel corpus of pseudo-tweets for english,\njapanese and chinese it can also be used to train translation models between\nthese three languages." ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"fol-00\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"fol-00\"\n\nMore Information needed" ]

[ "arxiv:2312.15915", "region:us" ]

[ "2312.15915" ]

[]

[ "# ChartBench: A Benchmark for Complex Visual Reasoning in Charts\n\n<a href='URL src='URL <a href='URL src='URL <a href='URL src='URL ![Dataset](URL", "## Introduction\n\nWe propose the challenging ChartBench to evaluate the chart recognition of MLLMs. \n!ChartBench Pipeline.\n\nWe improve the *Acc+* metric to avoid the randomly guessing situations.\n!improved Acc+ metric.\n\nWe collect a larger set of unlabeled charts to emphasize the MLLM's ability to interpret visual information without the aid of annotated data points.\n!Chart distributions and ChartCoT.", "## Todo\n- [ ] Open source all data of ChartBench.\n- [x] Open source the evaluate scripts.\n- [x] Open source the inference scripts.\n- [x] Open source the demo data (10%).", "## Setup\nPlease follow the official repository instructions below to set up the local environment.\n\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/llava_vl.URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL", "## Inference\n1. Complete the basic environment setup.\n2. Set 'task_name' in './Repos/URL', such as 'test' or 'BLIP2_Style'.\n3. Select or set the desired system prompt in './Repos/URL'.\n4. Modify the default path of 'CKPT_PATH' in './Repos/{MODEL_NAME}/URL'.\n5. Run 'URL' following the command format in './Scripts/URL'.\n6. The results are saved by default in './Eval/{task_name}/{MODEL_NAME}'.\n7. Set the parameters in './Scripts/stat_acc_plus.py' and the statistical results are saved in './Eval/{task_name}/Eval_Result'.", "## Ranking\n\n\n!ChartBench Pipeline." ]

[ "TAGS\n#arxiv-2312.15915 #region-us \n", "# ChartBench: A Benchmark for Complex Visual Reasoning in Charts\n\n<a href='URL src='URL <a href='URL src='URL <a href='URL src='URL ![Dataset](URL", "## Introduction\n\nWe propose the challenging ChartBench to evaluate the chart recognition of MLLMs. \n!ChartBench Pipeline.\n\nWe improve the *Acc+* metric to avoid the randomly guessing situations.\n!improved Acc+ metric.\n\nWe collect a larger set of unlabeled charts to emphasize the MLLM's ability to interpret visual information without the aid of annotated data points.\n!Chart distributions and ChartCoT.", "## Todo\n- [ ] Open source all data of ChartBench.\n- [x] Open source the evaluate scripts.\n- [x] Open source the inference scripts.\n- [x] Open source the demo data (10%).", "## Setup\nPlease follow the official repository instructions below to set up the local environment.\n\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/llava_vl.URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL\n- <a href='URL src='URL/URL", "## Inference\n1. Complete the basic environment setup.\n2. Set 'task_name' in './Repos/URL', such as 'test' or 'BLIP2_Style'.\n3. Select or set the desired system prompt in './Repos/URL'.\n4. Modify the default path of 'CKPT_PATH' in './Repos/{MODEL_NAME}/URL'.\n5. Run 'URL' following the command format in './Scripts/URL'.\n6. The results are saved by default in './Eval/{task_name}/{MODEL_NAME}'.\n7. Set the parameters in './Scripts/stat_acc_plus.py' and the statistical results are saved in './Eval/{task_name}/Eval_Result'.", "## Ranking\n\n\n!ChartBench Pipeline." ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"chonk-data\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"chonk-data\"\n\nMore Information needed" ]

[ "language:en", "region:us" ]

[]

[ "en" ]

[ "# Dataset Card for \"Bollywood_songs\"\n\nMore Information needed" ]

[ "TAGS\n#language-English #region-us \n", "# Dataset Card for \"Bollywood_songs\"\n\nMore Information needed" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for Evaluation run of SC99/Mistral-7B-summ-lora-tuned-8h\n\n\n\nDataset automatically created during the evaluation run of model SC99/Mistral-7B-summ-lora-tuned-8h on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-29T11:29:57.357503(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "TAGS\n#region-us \n", "# Dataset Card for Evaluation run of SC99/Mistral-7B-summ-lora-tuned-8h\n\n\n\nDataset automatically created during the evaluation run of model SC99/Mistral-7B-summ-lora-tuned-8h on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-29T11:29:57.357503(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"ultrafeedback-capybara-mix-5k\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"ultrafeedback-capybara-mix-5k\"\n\nMore Information needed" ]

[ "task_categories:text-generation", "size_categories:100K<n<1M", "license:bsd-3-clause", "fine-tuning", "function-calling", "instruct", "sft", "region:us" ]

[]

[]

[ "# Pandora Instruct\n\nAn instruction dataset for Supervised fine-tuning of the Pandora Large Language Model (LLM).\n\nThe dataset is based on the existing datasets:\n\n- ('teknium/openhermes')[URL\n- ('ise-uiuc/magicoder-evol-instruct-110k')[URL\n- ('ise-uiuc/magicoder-oss-instruct-75k')[URL\n- ('glaiveai/glaive-function-calling-v2')[URL", "## Copyright and license\n\nCopyright (c) 2024, Danilo Peixoto Ferreira. All rights reserved.\n\nProject developed under a BSD-3-Clause license." ]

[ "TAGS\n#task_categories-text-generation #size_categories-100K<n<1M #license-bsd-3-clause #fine-tuning #function-calling #instruct #sft #region-us \n", "# Pandora Instruct\n\nAn instruction dataset for Supervised fine-tuning of the Pandora Large Language Model (LLM).\n\nThe dataset is based on the existing datasets:\n\n- ('teknium/openhermes')[URL\n- ('ise-uiuc/magicoder-evol-instruct-110k')[URL\n- ('ise-uiuc/magicoder-oss-instruct-75k')[URL\n- ('glaiveai/glaive-function-calling-v2')[URL", "## Copyright and license\n\nCopyright (c) 2024, Danilo Peixoto Ferreira. All rights reserved.\n\nProject developed under a BSD-3-Clause license." ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"LogicNLI\"\n\n\n\nURL" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"LogicNLI\"\n\n\n\nURL" ]

[ "license:ecl-2.0", "region:us" ]

[]

[]

[]

[ "TAGS\n#license-ecl-2.0 #region-us \n" ]

[ "task_categories:object-detection", "size_categories:1K<n<10K", "license:mit", "industry", "doi:10.57967/hf/1710", "region:us" ]

[]

[]

[ "### Download the dataset:" ]

[ "TAGS\n#task_categories-object-detection #size_categories-1K<n<10K #license-mit #industry #doi-10.57967/hf/1710 #region-us \n", "### Download the dataset:" ]

[ "region:us" ]

[]

[]

[ "### Human: Who founded ABCDE Capital in 2022. ### Assistant: Hey! ABCDE Capital is founded by Du Jun and BMAN in 2022.\"}" ]

[ "TAGS\n#region-us \n", "### Human: Who founded ABCDE Capital in 2022. ### Assistant: Hey! ABCDE Capital is founded by Du Jun and BMAN in 2022.\"}" ]

[ "region:us" ]

[]

[]

[]

[ "TAGS\n#region-us \n" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for Evaluation run of SC99/Mistral-7B-summ-ia3-tuned-8h\n\n\n\nDataset automatically created during the evaluation run of model SC99/Mistral-7B-summ-ia3-tuned-8h on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-29T13:30:16.956785(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "TAGS\n#region-us \n", "# Dataset Card for Evaluation run of SC99/Mistral-7B-summ-ia3-tuned-8h\n\n\n\nDataset automatically created during the evaluation run of model SC99/Mistral-7B-summ-ia3-tuned-8h on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-29T13:30:16.956785(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "license:lgpl", "region:us" ]

[]

[]

[ "# Dataset Card for Dataset Name\n\nThis is a complement to the TREC CaST (2020-22) datasets, with pre-computed offset relative to the original files." ]

[ "TAGS\n#license-lgpl #region-us \n", "# Dataset Card for Dataset Name\n\nThis is a complement to the TREC CaST (2020-22) datasets, with pre-computed offset relative to the original files." ]

[ "task_categories:text-generation", "language:de", "region:us" ]

[]

[ "de" ]

[]

[ "TAGS\n#task_categories-text-generation #language-German #region-us \n" ]

[ "task_categories:text-generation", "language:de", "region:us" ]

[]

[ "de" ]

[]

[ "TAGS\n#task_categories-text-generation #language-German #region-us \n" ]

[ "task_categories:text-generation", "language:de", "region:us" ]

[]

[ "de" ]

[]

[ "TAGS\n#task_categories-text-generation #language-German #region-us \n" ]

[ "task_categories:text-generation", "language:de", "region:us" ]

[]

[ "de" ]

[]

[ "TAGS\n#task_categories-text-generation #language-German #region-us \n" ]

[ "task_categories:text-generation", "language:de", "region:us" ]

[]

[ "de" ]

[]

[ "TAGS\n#task_categories-text-generation #language-German #region-us \n" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for Evaluation run of Qwen/Qwen2-beta-14B\n\n\n\nDataset automatically created during the evaluation run of model Qwen/Qwen2-beta-14B on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-30T11:39:41.356084(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "TAGS\n#region-us \n", "# Dataset Card for Evaluation run of Qwen/Qwen2-beta-14B\n\n\n\nDataset automatically created during the evaluation run of model Qwen/Qwen2-beta-14B on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-30T11:39:41.356084(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "task_categories:conversational", "task_categories:question-answering", "size_categories:10K<n<100K", "language:en", "language:zh", "Long Context", "sft", "arxiv:2401.18058", "region:us" ]

[ "2401.18058" ]

[ "en", "zh" ]

[]

[ "TAGS\n#task_categories-conversational #task_categories-question-answering #size_categories-10K<n<100K #language-English #language-Chinese #Long Context #sft #arxiv-2401.18058 #region-us \n" ]

[ "language:en", "license:cc-by-nc-4.0", "not-for-all-audiences", "region:us" ]

[]

[ "en" ]

[]

[ "TAGS\n#language-English #license-cc-by-nc-4.0 #not-for-all-audiences #region-us \n" ]

[ "task_categories:text-generation", "language:de", "region:us" ]

[]

[ "de" ]

[]

[ "TAGS\n#task_categories-text-generation #language-German #region-us \n" ]

[ "license:cc-by-4.0", "arxiv:2401.15847", "region:us" ]

[ "2401.15847" ]

[]

[]

[ "TAGS\n#license-cc-by-4.0 #arxiv-2401.15847 #region-us \n" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"quesst14_all_unit\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"quesst14_all_unit\"\n\nMore Information needed" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"VietnameseMedicineQA-raw-parquet\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"VietnameseMedicineQA-raw-parquet\"\n\nMore Information needed" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"hf-codegen_v1\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"hf-codegen_v1\"\n\nMore Information needed" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"cowese_multiplechoice_top2vec\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"cowese_multiplechoice_top2vec\"\n\nMore Information needed" ]

[ "license:cc-by-4.0", "region:us" ]

[]

[]

[]

[ "TAGS\n#license-cc-by-4.0 #region-us \n" ]

[ "task_categories:text-generation", "size_categories:100K<n<1M", "code", "python", "region:us" ]

[]

[]

[]

[ "TAGS\n#task_categories-text-generation #size_categories-100K<n<1M #code #python #region-us \n" ]

[ "task_categories:question-answering", "size_categories:1K<n<10K", "language:en", "license:mit", "region:us" ]

[]

[ "en" ]

[ "# Dataset Card for \"utility_qa\"\n\nMore Information needed" ]

[ "TAGS\n#task_categories-question-answering #size_categories-1K<n<10K #language-English #license-mit #region-us \n", "# Dataset Card for \"utility_qa\"\n\nMore Information needed" ]

[ "task_categories:text-generation", "language:zh", "license:apache-2.0", "region:us" ]

[]

[ "zh" ]

[ "### 现有角色扮演数据集", "### 我们的优势\n\n\n* 精细化的角色设定数据：我们的数据集弥补了现有开源Role-playing Instruction数据中常见的角色设定不足问题。我们从角色身份、语言风格、背景故事等多个维度提供了详尽的信息。特别是在GPT-4版本中，我们还增加了对话者身份的设定，使数据更为完整和丰富。\n* 多样性的角色性格：本数据集涵盖尽可能广泛的二次元角色性格，保证了低重复性和高丰富度。\n* 生动的语言和动作描述：我们的数据集不仅包含角色间的对话，还添加了角色的动作描述，使得对话更加生动和真实，将为用户提供更丰富的角色扮演体验。\n* 通用角色扮演数据生成框架：我们提供了一个通用的角色扮演数据生成框架，充分释放OpenAI API的角色扮演能力。该框架生成的数据将用于微调和RAG。目前，该框架代码正在进行测试和优化，预计将在不久的将来公开。\n\n\n联系我们\n----\n\n\n如有需要或任何疑问请联系：邮箱：info@URL\n\n\n项目使用与免责声明\n---------\n\n\n本项目遵循Apache 2.0许可协议。在此协议下，您被授权自由使用项目中的代码进行商业活动。然而，若本项目涉及到特定角色的版权问题，或受其他相关协议限制（例如接口使用协议等），您使用时必须严格遵守这些协议的相关条款。\n\n\n本项目所开源的数据是通过调用OpenAI接口生成的，并未经过严格的事实和安全性验证。因此，在使用这些数据时，请您务必谨慎考虑其真实性、准确性以及安全性。同时，请确保在使用过程中遵守OpenAI的相关规定。\n\n\n此外，我们声明，本数据集不代表开发者或任何其他方的立场、利益或观点，也不代表任何团体的主张。本项目的开发者不对使用本数据集可能引起的任何形式的损害或纠纷承担责任。" ]

[ "TAGS\n#task_categories-text-generation #language-Chinese #license-apache-2.0 #region-us \n", "### 现有角色扮演数据集", "### 我们的优势\n\n\n* 精细化的角色设定数据：我们的数据集弥补了现有开源Role-playing Instruction数据中常见的角色设定不足问题。我们从角色身份、语言风格、背景故事等多个维度提供了详尽的信息。特别是在GPT-4版本中，我们还增加了对话者身份的设定，使数据更为完整和丰富。\n* 多样性的角色性格：本数据集涵盖尽可能广泛的二次元角色性格，保证了低重复性和高丰富度。\n* 生动的语言和动作描述：我们的数据集不仅包含角色间的对话，还添加了角色的动作描述，使得对话更加生动和真实，将为用户提供更丰富的角色扮演体验。\n* 通用角色扮演数据生成框架：我们提供了一个通用的角色扮演数据生成框架，充分释放OpenAI API的角色扮演能力。该框架生成的数据将用于微调和RAG。目前，该框架代码正在进行测试和优化，预计将在不久的将来公开。\n\n\n联系我们\n----\n\n\n如有需要或任何疑问请联系：邮箱：info@URL\n\n\n项目使用与免责声明\n---------\n\n\n本项目遵循Apache 2.0许可协议。在此协议下，您被授权自由使用项目中的代码进行商业活动。然而，若本项目涉及到特定角色的版权问题，或受其他相关协议限制（例如接口使用协议等），您使用时必须严格遵守这些协议的相关条款。\n\n\n本项目所开源的数据是通过调用OpenAI接口生成的，并未经过严格的事实和安全性验证。因此，在使用这些数据时，请您务必谨慎考虑其真实性、准确性以及安全性。同时，请确保在使用过程中遵守OpenAI的相关规定。\n\n\n此外，我们声明，本数据集不代表开发者或任何其他方的立场、利益或观点，也不代表任何团体的主张。本项目的开发者不对使用本数据集可能引起的任何形式的损害或纠纷承担责任。" ]

[ "license:mit", "region:us" ]

[]

[]

[]

[ "TAGS\n#license-mit #region-us \n" ]

[ "task_categories:text-generation", "task_ids:language-modeling", "annotations_creators:crowdsourced", "language_creators:crowdsourced", "multilinguality:monolingual", "size_categories:1M<n<10M", "source_datasets:original", "language:ru", "license:cc0-1.0", "region:us" ]

[]

[ "ru" ]

[ "# Dataset Card for Kazanexpress products", "### Dataset Summary\n\nThis dataset was scraped from product pages on the Russian marketplace Kazanexpress. It includes all information from the product card and metadata from the API. The dataset was collected by processing around 3 million products, starting from the first one. At the time the dataset was collected, it is assumed that these were all the products available on this marketplace. Please note that the data returned by the API was not processed. This means that there is a large amount of extraneous data in the dataset, such as links to images in descriptions and parts of HTML. Additionally, some fields may be empty, but the string is expected to contain some data, empty responses have been sorted.", "### Languages\n\nThe dataset is mostly in Russian, but there may be other languages present.", "## Dataset Structure", "### Data Fields\n\nThis dataset includes the following fields:\n\n- 'id': Identifier for the product (integer)\n- 'title': Title of the product (string)\n- 'category': Category of the product (string)\n- 'parent_category': Parent category of the product (string)\n- 'grandparent_category': Grandparent category of the product (string)\n- 'greatgrandparent_category': Great-grandparent category of the product (string)\n- 'rating': Rating of the product (float)\n- 'description': Description of the product (string)\n- 'comments': Comments about the product (array)\n- 'attributes': Attributes of the product (array)\n- 'seller': Name of the seller (string)\n- 'seller_description': Description provided by the seller (string)\n- 'adultCategory': Whether the product is for adults (boolean)", "### Data Splits\n\nAll examples are in the train split, there is no validation split.", "## Additional Information", "### License\n\nThis dataset is dedicated to the public domain under the Creative Commons Zero (CC0) license. This means you can:\n\n* Use it for any purpose, including commercial projects.\n* Modify it however you like.\n* Distribute it without asking permission.\n\nNo attribution is required, but it's always appreciated!\n\nCC0 license: URL\n\nTo learn more about CC0, visit the Creative Commons website: URL", "### Dataset Curators\n\n- nyuuzyou" ]

[ "TAGS\n#task_categories-text-generation #task_ids-language-modeling #annotations_creators-crowdsourced #language_creators-crowdsourced #multilinguality-monolingual #size_categories-1M<n<10M #source_datasets-original #language-Russian #license-cc0-1.0 #region-us \n", "# Dataset Card for Kazanexpress products", "### Dataset Summary\n\nThis dataset was scraped from product pages on the Russian marketplace Kazanexpress. It includes all information from the product card and metadata from the API. The dataset was collected by processing around 3 million products, starting from the first one. At the time the dataset was collected, it is assumed that these were all the products available on this marketplace. Please note that the data returned by the API was not processed. This means that there is a large amount of extraneous data in the dataset, such as links to images in descriptions and parts of HTML. Additionally, some fields may be empty, but the string is expected to contain some data, empty responses have been sorted.", "### Languages\n\nThe dataset is mostly in Russian, but there may be other languages present.", "## Dataset Structure", "### Data Fields\n\nThis dataset includes the following fields:\n\n- 'id': Identifier for the product (integer)\n- 'title': Title of the product (string)\n- 'category': Category of the product (string)\n- 'parent_category': Parent category of the product (string)\n- 'grandparent_category': Grandparent category of the product (string)\n- 'greatgrandparent_category': Great-grandparent category of the product (string)\n- 'rating': Rating of the product (float)\n- 'description': Description of the product (string)\n- 'comments': Comments about the product (array)\n- 'attributes': Attributes of the product (array)\n- 'seller': Name of the seller (string)\n- 'seller_description': Description provided by the seller (string)\n- 'adultCategory': Whether the product is for adults (boolean)", "### Data Splits\n\nAll examples are in the train split, there is no validation split.", "## Additional Information", "### License\n\nThis dataset is dedicated to the public domain under the Creative Commons Zero (CC0) license. This means you can:\n\n* Use it for any purpose, including commercial projects.\n* Modify it however you like.\n* Distribute it without asking permission.\n\nNo attribution is required, but it's always appreciated!\n\nCC0 license: URL\n\nTo learn more about CC0, visit the Creative Commons website: URL", "### Dataset Curators\n\n- nyuuzyou" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"ultrachat_sft_instruction_format\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"ultrachat_sft_instruction_format\"\n\nMore Information needed" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"ultrachat_gen_instruction_format\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"ultrachat_gen_instruction_format\"\n\nMore Information needed" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"ultrafeedback_binarized_narrow_unlabeled\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"ultrafeedback_binarized_narrow_unlabeled\"\n\nMore Information needed" ]

[ "task_categories:text2text-generation", "size_categories:1K<n<10K", "language:en", "license:mit", "text-to-sql", "arxiv:2106.01065", "region:us" ]

[ "2106.01065" ]

[ "en" ]

[ "# Dataset Card for Sypder-Syn\n\nSpyder-Syn is a human curated variant of the Spider Text-to-SQL database.\nThe database was created to test the robustness of text-to-SQL models for robustness of synonym substitution.\n\nThe source GIT repo for Sypder-Syn is located here: URL\n\nDetails regarding the data perterbation methods used and objectives are described in ACL 2021: arXiv", "## Paper Abstract\n\n> Recently, there has been significant progress in studying neural networks to translate text descriptions into SQL queries. Despite achieving good performance on some public benchmarks, existing text-to-SQL models typically rely on the lexical matching between words in natural language (NL) questions and tokens in table schemas, which may render the models vulnerable to attacks that break the schema linking mechanism. In this work, we investigate the robustness of text-to-SQL models to synonym substitution. In particular, we introduce Spider-Syn, a human-curated dataset based on the Spider benchmark for text-to-SQL translation. NL questions in Spider-Syn are modified from Spider, by replacing their schema-related words with manually selected synonyms that reflect real-world question paraphrases. We observe that the accuracy dramatically drops by eliminating such explicit correspondence between NL questions and table schemas, even if the synonyms are not adversarially selected to conduct worst-case adversarial attacks. Finally, we present two categories of approaches to improve the model robustness. The first category of approaches utilizes additional synonym annotations for table schemas by modifying the model input, while the second category is based on adversarial training. We demonstrate that both categories of approaches significantly outperform their counterparts without the defense, and the first category of approaches are more effective." ]

[ "TAGS\n#task_categories-text2text-generation #size_categories-1K<n<10K #language-English #license-mit #text-to-sql #arxiv-2106.01065 #region-us \n", "# Dataset Card for Sypder-Syn\n\nSpyder-Syn is a human curated variant of the Spider Text-to-SQL database.\nThe database was created to test the robustness of text-to-SQL models for robustness of synonym substitution.\n\nThe source GIT repo for Sypder-Syn is located here: URL\n\nDetails regarding the data perterbation methods used and objectives are described in ACL 2021: arXiv", "## Paper Abstract\n\n> Recently, there has been significant progress in studying neural networks to translate text descriptions into SQL queries. Despite achieving good performance on some public benchmarks, existing text-to-SQL models typically rely on the lexical matching between words in natural language (NL) questions and tokens in table schemas, which may render the models vulnerable to attacks that break the schema linking mechanism. In this work, we investigate the robustness of text-to-SQL models to synonym substitution. In particular, we introduce Spider-Syn, a human-curated dataset based on the Spider benchmark for text-to-SQL translation. NL questions in Spider-Syn are modified from Spider, by replacing their schema-related words with manually selected synonyms that reflect real-world question paraphrases. We observe that the accuracy dramatically drops by eliminating such explicit correspondence between NL questions and table schemas, even if the synonyms are not adversarially selected to conduct worst-case adversarial attacks. Finally, we present two categories of approaches to improve the model robustness. The first category of approaches utilizes additional synonym annotations for table schemas by modifying the model input, while the second category is based on adversarial training. We demonstrate that both categories of approaches significantly outperform their counterparts without the defense, and the first category of approaches are more effective." ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"HumanPoseTest\"", "## Source\n\nDataset created from human activitiy recognition", "## Task\n\nFind the image (from image2, image3, and image4) that has the same marked body part as in image1. The 18 possible body parts (e.g., nose, right ankle, left elbow, neck, left shoulder, etc.) are marked with green dots.\n - image2: A\n - image3: B\n - image4: C", "## Prompt\nConcise version:\n\n\nVerbose version:" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"HumanPoseTest\"", "## Source\n\nDataset created from human activitiy recognition", "## Task\n\nFind the image (from image2, image3, and image4) that has the same marked body part as in image1. The 18 possible body parts (e.g., nose, right ankle, left elbow, neck, left shoulder, etc.) are marked with green dots.\n - image2: A\n - image3: B\n - image4: C", "## Prompt\nConcise version:\n\n\nVerbose version:" ]

[ "license:odbl", "region:us" ]

[]

[]

[ "# Dataset Card for \"HumanPose\"", "## Source\n\nDataset created from human activitiy recognition", "## Task\n\nFind the image (from image2, image3, and image4) that has the same marked body part as in image1. The 18 possible body parts (e.g., nose, right ankle, left elbow, neck, left shoulder, etc.) are marked with green dots.\n - image2: A\n - image3: B\n - image4: C\n - cannot tell: D", "## Prompt" ]

[ "TAGS\n#license-odbl #region-us \n", "# Dataset Card for \"HumanPose\"", "## Source\n\nDataset created from human activitiy recognition", "## Task\n\nFind the image (from image2, image3, and image4) that has the same marked body part as in image1. The 18 possible body parts (e.g., nose, right ankle, left elbow, neck, left shoulder, etc.) are marked with green dots.\n - image2: A\n - image3: B\n - image4: C\n - cannot tell: D", "## Prompt" ]

[ "region:us" ]

[]

[]

[ "## Source\ndataset created from (Practice) Graphical Reasoning Questions for Civil Service Exam in China.", "## Task\nIdentify the one picture that follows the same pattern or rule established by the previous pictures.", "## Prompt:" ]

[ "TAGS\n#region-us \n", "## Source\ndataset created from (Practice) Graphical Reasoning Questions for Civil Service Exam in China.", "## Task\nIdentify the one picture that follows the same pattern or rule established by the previous pictures.", "## Prompt:" ]

[ "region:us" ]

[]

[]

[ "## Source\ndataset created from (Practice) Graphical Reasoning Questions for Civil Service Exam in China.", "## Task\nIdentify the one picture that follows the same pattern or rule established by the previous pictures.", "## Prompt:" ]

[ "TAGS\n#region-us \n", "## Source\ndataset created from (Practice) Graphical Reasoning Questions for Civil Service Exam in China.", "## Task\nIdentify the one picture that follows the same pattern or rule established by the previous pictures.", "## Prompt:" ]

[ "whisper", "whisperkit", "coreml", "asr", "quantized", "region:us" ]

[]

[]

[ "### WhisperKit + 'openai\\_whisper-large-v3' (+optimized variants)", "### Different Projects + 'openai\\_whisper-large-v3'", "### Quality-of-Inference (QoI) Certification\n\n\nWe believe that rigorously measuring the quality of inference is necessary for developers and\nenterprises to make informed decisions when opting to use optimized or compressed variants of\nWhisper models in production. The current measurements are between reference and optimized\nWhisperKit models. We are going to extend the scope of this measurement to other Whisper\nimplementations soon so developers can certify the behavior change (if any) caused by\nalternating use of WhisperKit with (or migration from) these implementations.\n\n\nIn all measurements, we care primarily about per-example no-regressions (quantified as 'qoi' below)\nwhich is a stricter metric compared to dataset average WER. A 100% 'qoi' preserves perfect\nbackwards-compatibility on the test distribution and avoids \"perceived regressions\", the phenomenon\nwhere per-example known behavior changes after a code/model update and causes divergence in\ndownstream code or breaks the user experience itself (even if dataset averages might stay flat\nacross updates). Pseudocode for 'qoi':\n\n\nWe define the reference model as the default float16 precision Core ML model that is generated by\nwhisperkittools. This reference model matches the accuracy of the original PyTorch model\non the specified test sets. We use 'librispeech/URL' (5 hours of short English audio clips)\nas our testing set for Whisper. We are actively expanding our test set coverage to 'earnings22'\n(120 hours of long English audio clips with various accents). We anticipate developers that use Whisper in production to have\ntheir own Quality Assurance test sets and whisperkittools offers the tooling necessary to run the\nsame measurements on such custom test sets, please see the Model Evaluation on Custom Dataset\nfor details.", "### Reproducing Results\n\n\nResults in this page are generated by our cluster of Apple Silicon Macs. We use them as self-hosted runners on\nGithub Actions as our CI infrastructure. Due to security concerns,\nwe are unable to open up the cluster to the public. However, any Apple Silicon Mac (even with 8GB RAM) can be used to\nrun identical evaluation jobs\nlocally. For reference, our M2 Ultra devices complete a 'librispeech' + 'openai/whisper-large-v3'\nevaluation in under 1 hour regardless of the Whisper implementation. Older Apple Silicon Macs should take less than\n1 day to complete the same evaluation.\n\n\nGlossary:\n\n\n* '\\_turbo': Indicates the presence of additional optimizations (not compression) to unlock streaming transcription\nas described in our Blog Post.\n* '\\_\\*MB': Indicates the presence of mixed-bit quantization. Instead of cluttering the filename with details like\n'\\_AudioEncoder-5.8bits\\_TextDecoder-6.1bits', we choose to summarize the compression spec as the resulting total file size since this is what matters to developers in production." ]

[ "TAGS\n#whisper #whisperkit #coreml #asr #quantized #region-us \n", "### WhisperKit + 'openai\\_whisper-large-v3' (+optimized variants)", "### Different Projects + 'openai\\_whisper-large-v3'", "### Quality-of-Inference (QoI) Certification\n\n\nWe believe that rigorously measuring the quality of inference is necessary for developers and\nenterprises to make informed decisions when opting to use optimized or compressed variants of\nWhisper models in production. The current measurements are between reference and optimized\nWhisperKit models. We are going to extend the scope of this measurement to other Whisper\nimplementations soon so developers can certify the behavior change (if any) caused by\nalternating use of WhisperKit with (or migration from) these implementations.\n\n\nIn all measurements, we care primarily about per-example no-regressions (quantified as 'qoi' below)\nwhich is a stricter metric compared to dataset average WER. A 100% 'qoi' preserves perfect\nbackwards-compatibility on the test distribution and avoids \"perceived regressions\", the phenomenon\nwhere per-example known behavior changes after a code/model update and causes divergence in\ndownstream code or breaks the user experience itself (even if dataset averages might stay flat\nacross updates). Pseudocode for 'qoi':\n\n\nWe define the reference model as the default float16 precision Core ML model that is generated by\nwhisperkittools. This reference model matches the accuracy of the original PyTorch model\non the specified test sets. We use 'librispeech/URL' (5 hours of short English audio clips)\nas our testing set for Whisper. We are actively expanding our test set coverage to 'earnings22'\n(120 hours of long English audio clips with various accents). We anticipate developers that use Whisper in production to have\ntheir own Quality Assurance test sets and whisperkittools offers the tooling necessary to run the\nsame measurements on such custom test sets, please see the Model Evaluation on Custom Dataset\nfor details.", "### Reproducing Results\n\n\nResults in this page are generated by our cluster of Apple Silicon Macs. We use them as self-hosted runners on\nGithub Actions as our CI infrastructure. Due to security concerns,\nwe are unable to open up the cluster to the public. However, any Apple Silicon Mac (even with 8GB RAM) can be used to\nrun identical evaluation jobs\nlocally. For reference, our M2 Ultra devices complete a 'librispeech' + 'openai/whisper-large-v3'\nevaluation in under 1 hour regardless of the Whisper implementation. Older Apple Silicon Macs should take less than\n1 day to complete the same evaluation.\n\n\nGlossary:\n\n\n* '\\_turbo': Indicates the presence of additional optimizations (not compression) to unlock streaming transcription\nas described in our Blog Post.\n* '\\_\\*MB': Indicates the presence of mixed-bit quantization. Instead of cluttering the filename with details like\n'\\_AudioEncoder-5.8bits\\_TextDecoder-6.1bits', we choose to summarize the compression spec as the resulting total file size since this is what matters to developers in production." ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for Evaluation run of Qwen/Qwen2-beta-72B\n\n\n\nDataset automatically created during the evaluation run of model Qwen/Qwen2-beta-72B on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-29T23:47:04.571636(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "TAGS\n#region-us \n", "# Dataset Card for Evaluation run of Qwen/Qwen2-beta-72B\n\n\n\nDataset automatically created during the evaluation run of model Qwen/Qwen2-beta-72B on the Open LLM Leaderboard.\n\nThe dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.\n\nThe dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The \"train\" split is always pointing to the latest results.\n\nAn additional configuration \"results\" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the Open LLM Leaderboard).\n\nTo load the details from a run, you can for instance do the following:", "## Latest results\n\nThese are the latest results from run 2024-01-29T23:47:04.571636(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the \"latest\" split for each eval):", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"relabeled_alpacafarm_pythiasft_20K_preference_data_modelength\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"relabeled_alpacafarm_pythiasft_20K_preference_data_modelength\"\n\nMore Information needed" ]

[ "task_categories:translation", "size_categories:10M<n<100M", "language:ja", "language:ko", "license:cc-by-nc-sa-4.0", "region:us" ]

[]

[ "ja", "ko" ]

[ "# Dataset Card for Dataset Name\n\nThis dataset contains Japanese-Korean paired text which is from Helsinki-NLP/Tatoeba-Challenge.", "## Dataset Details", "### Dataset Description\n\n- Curated by: Helsinki-NLP\n- Language(s) (NLP): Japanese-Korean\n- License: CC BY-NC-SA 4.0", "### Dataset Sources\n\n- Repository: Helsinki-NLP/Tatoeba-Challenge\n- Detail: Japanese - Korean jpn-kor", "## Uses\n\nThe dataset can be used to train the translation model that translates Japanese sentence to Korean.", "### Out-of-Scope Use\n\nYou cannot use this dataset to train the model which is to be used under commercial service.", "## Dataset Structure\n\n\n\nEach dataset has two columns: 'sourceString' and 'targetString', which corresponds to Japanese and Korean sentence. \nCheck example code to learn how to load the dataset.", "## Dataset Creation", "### Personal and Sensitive Information\n\n\n\nThis dataset may contain following inappropriate or explicit sentences:\n- personal\n- sensitive\n- private\n - data that reveals addresses\n - uniquely identifiable names or aliases\n - racial or ethnic origins\n - sexual orientations\n - religious beliefs\n - political opinions\n - financial or health data\n - etc.\n\nSo use with your own risk.\n\n\nBibTeX:", "## Dataset Card Authors\n\nsappho192", "## Dataset Card Contact\n\nPlease create a thread in the community." ]

[ "TAGS\n#task_categories-translation #size_categories-10M<n<100M #language-Japanese #language-Korean #license-cc-by-nc-sa-4.0 #region-us \n", "# Dataset Card for Dataset Name\n\nThis dataset contains Japanese-Korean paired text which is from Helsinki-NLP/Tatoeba-Challenge.", "## Dataset Details", "### Dataset Description\n\n- Curated by: Helsinki-NLP\n- Language(s) (NLP): Japanese-Korean\n- License: CC BY-NC-SA 4.0", "### Dataset Sources\n\n- Repository: Helsinki-NLP/Tatoeba-Challenge\n- Detail: Japanese - Korean jpn-kor", "## Uses\n\nThe dataset can be used to train the translation model that translates Japanese sentence to Korean.", "### Out-of-Scope Use\n\nYou cannot use this dataset to train the model which is to be used under commercial service.", "## Dataset Structure\n\n\n\nEach dataset has two columns: 'sourceString' and 'targetString', which corresponds to Japanese and Korean sentence. \nCheck example code to learn how to load the dataset.", "## Dataset Creation", "### Personal and Sensitive Information\n\n\n\nThis dataset may contain following inappropriate or explicit sentences:\n- personal\n- sensitive\n- private\n - data that reveals addresses\n - uniquely identifiable names or aliases\n - racial or ethnic origins\n - sexual orientations\n - religious beliefs\n - political opinions\n - financial or health data\n - etc.\n\nSo use with your own risk.\n\n\nBibTeX:", "## Dataset Card Authors\n\nsappho192", "## Dataset Card Contact\n\nPlease create a thread in the community." ]

[ "language:en", "license:cc-by-4.0", "medical", "MRI", "spine", "segmentation", "arxiv:2306.12217", "region:us" ]

[ "2306.12217" ]

[ "en" ]

[ "# Dataset Card for Spine Segmentation: Discs, Vertebrae and Spinal Canal (SPIDER)\n\nThe SPIDER data set contains lumbar spine magnetic resonance images (MRI) and segmentation masks described in the following paper:\n\nJasper W. van der Graaf, Miranda L. van Hooff, Constantinus F. M. Buckens, Matthieu Rutten, \nJob L. C. van Susante, Robert Jan Kroeze, Marinus de Kleuver, Bram van Ginneken, Nikolas Lessmann. (2023).\n*Lumbar spine segmentation in MR images: a dataset and a public benchmark.* URL\n\nThe data were made publicly available through Zenodo, an open repository operated by CERN, and posted on \nGrand Challenge.\n\n(*Disclaimer: I am not affiliated in any way with the aforementioned paper, researchers, or organizations. My only contribution is to curate the SPIDER data set \nhere on Hugging Face to increase accessibility. While I have taken care to curate the data in a way that maintains the integrity of the original data, any findings using this\nparticular data set should be validated against the original data provided by the researchers on Zenodo.*)", "## Table of Contents\n\n(Placeholder--to be filled in at end of project)", "## Dataset Description\n\n- Paper: Lumbar spine segmentation in MR images: a dataset and a public benchmark\n- Repository: Zenodo", "### Dataset Summary\n\nThe dataset includes 447 sagittal T1 and T2 MRI series collected from 218 patients across four hospitals.\nSegmentation masks indicating the vertebrae, intervertebral discs (IVDs), and spinal canal are also included.\nSegmentation masks were created manually by a medical trainee under the supervision of\na medical imaging expert and an experienced musculoskeletal radiologist.\n\nIn addition to MR images and segmentation masks, additional metadata (e.g., scanner manufacturer, pixel bandwidth, etc.), limited\npatient characteristics (biological sex and age, when available), and radiological gradings indicating specific degenerative\nchanges can be loaded with the corresponding image data.", "## Dataset Structure", "### Data Instances\n\nThere are 447 images and corresponding segmentation masks for 218 unique patients.", "### Data Fields\n\nThe following list includes the data fields available for importing:\n\n- Numeric representation of image\n \n- Numeric representation of segmentation mask\n - vertebrae\n - intervertebral discs\n - spinal canal\n \n- Image characteristics\n - number of vertebrae\n - number of discs\n \n- Patient characteristics\n - biological sex\n - age\n \n- Scanner characteristics\n - manufacturer\n - manufacturer model name\n - serial number\n - software version\n - echo numbers\n - echo time\n - echo train length\n - flip angle\n - imaged nucleus\n - imaging frequency\n - inplane phase encoding direction\n - MR acquisition type\n - magnetic field strength\n - number of phase encoding steps\n - percent phase field of view\n - percent sampling\n - photometric interpretation\n - pixel bandwidth\n - pixel spacing\n - repetition time\n - specific absorption rate (SAR)\n - samples per pixel\n - scanning sequence\n - sequence name\n - series description\n - slice thickness\n - spacing between slices\n - specific character set\n - transmit coil name\n - window center\n - window width\n\n(TODO: Will add variable descriptions after proposal approval)", "### Data Splits\n\nThe training set contains [x] images distributed as follows:\n\n- Unique individuals: [x]\n\n- Standard sagittal T1 images: [x]\n- Standard sagittal T2 images: [y]\n- Standard sagittal T2 SPACE images: [z]\n- \nThe validation set contains 87 images distributed as follows:\n\n- Unique individuals: [x]\n\n- Standard sagittal T1 images: [x]\n- Standard sagittal T2 images: [y]\n- Standard sagittal T2 SPACE images: [z]\n\nAn additional hidden test set (not available through Hugging Face) is available on the SPIDER Grand Challenge.", "## Image Resolution\n\nStandard sagittal T1 and T2 image resolution ranges from 3.3 x 0.33 x 0.33 mm to 4.8 x 0.90 x 0.90 mm. \nSagittal T2 SPACE sequence images had a near isotropic spatial resolution with a voxel size of 0.90 x 0.47 x 0.47 mm.\nSource", "## Dataset Curation\n\nThe data have been curated to enable users to load any of the following:\n\n- Raw image files\n- Raw segmentation masks\n- Numeric representations of images in tensor format\n- Numeric representations of segmentation masks in tensor format\n- Linked patient characteristics (limited to sex and age, if available)\n- Linked scanner characteristics", "### Source Data", "### Processing Steps\n\n(Specifics to be determined, but will include:)\n\n1. Conversion of .mha files to numeric representations\n2. Linking of segmentation mask numeric representations to image files\n3. Linking of patient and scanner characteristics to image files\n4. Cleaning of patient and scanner characteristics", "## Additional Information", "### License\n\nThe dataset is published under a CC-BY 4.0 license: URL\n\nJasper W. van der Graaf, Miranda L. van Hooff, Constantinus F. M. Buckens, Matthieu Rutten, \nJob L. C. van Susante, Robert Jan Kroeze, Marinus de Kleuver, Bram van Ginneken, Nikolas Lessmann. (2023).\n*Lumbar spine segmentation in MR images: a dataset and a public benchmark.* URL" ]

[ "TAGS\n#language-English #license-cc-by-4.0 #medical #MRI #spine #segmentation #arxiv-2306.12217 #region-us \n", "# Dataset Card for Spine Segmentation: Discs, Vertebrae and Spinal Canal (SPIDER)\n\nThe SPIDER data set contains lumbar spine magnetic resonance images (MRI) and segmentation masks described in the following paper:\n\nJasper W. van der Graaf, Miranda L. van Hooff, Constantinus F. M. Buckens, Matthieu Rutten, \nJob L. C. van Susante, Robert Jan Kroeze, Marinus de Kleuver, Bram van Ginneken, Nikolas Lessmann. (2023).\n*Lumbar spine segmentation in MR images: a dataset and a public benchmark.* URL\n\nThe data were made publicly available through Zenodo, an open repository operated by CERN, and posted on \nGrand Challenge.\n\n(*Disclaimer: I am not affiliated in any way with the aforementioned paper, researchers, or organizations. My only contribution is to curate the SPIDER data set \nhere on Hugging Face to increase accessibility. While I have taken care to curate the data in a way that maintains the integrity of the original data, any findings using this\nparticular data set should be validated against the original data provided by the researchers on Zenodo.*)", "## Table of Contents\n\n(Placeholder--to be filled in at end of project)", "## Dataset Description\n\n- Paper: Lumbar spine segmentation in MR images: a dataset and a public benchmark\n- Repository: Zenodo", "### Dataset Summary\n\nThe dataset includes 447 sagittal T1 and T2 MRI series collected from 218 patients across four hospitals.\nSegmentation masks indicating the vertebrae, intervertebral discs (IVDs), and spinal canal are also included.\nSegmentation masks were created manually by a medical trainee under the supervision of\na medical imaging expert and an experienced musculoskeletal radiologist.\n\nIn addition to MR images and segmentation masks, additional metadata (e.g., scanner manufacturer, pixel bandwidth, etc.), limited\npatient characteristics (biological sex and age, when available), and radiological gradings indicating specific degenerative\nchanges can be loaded with the corresponding image data.", "## Dataset Structure", "### Data Instances\n\nThere are 447 images and corresponding segmentation masks for 218 unique patients.", "### Data Fields\n\nThe following list includes the data fields available for importing:\n\n- Numeric representation of image\n \n- Numeric representation of segmentation mask\n - vertebrae\n - intervertebral discs\n - spinal canal\n \n- Image characteristics\n - number of vertebrae\n - number of discs\n \n- Patient characteristics\n - biological sex\n - age\n \n- Scanner characteristics\n - manufacturer\n - manufacturer model name\n - serial number\n - software version\n - echo numbers\n - echo time\n - echo train length\n - flip angle\n - imaged nucleus\n - imaging frequency\n - inplane phase encoding direction\n - MR acquisition type\n - magnetic field strength\n - number of phase encoding steps\n - percent phase field of view\n - percent sampling\n - photometric interpretation\n - pixel bandwidth\n - pixel spacing\n - repetition time\n - specific absorption rate (SAR)\n - samples per pixel\n - scanning sequence\n - sequence name\n - series description\n - slice thickness\n - spacing between slices\n - specific character set\n - transmit coil name\n - window center\n - window width\n\n(TODO: Will add variable descriptions after proposal approval)", "### Data Splits\n\nThe training set contains [x] images distributed as follows:\n\n- Unique individuals: [x]\n\n- Standard sagittal T1 images: [x]\n- Standard sagittal T2 images: [y]\n- Standard sagittal T2 SPACE images: [z]\n- \nThe validation set contains 87 images distributed as follows:\n\n- Unique individuals: [x]\n\n- Standard sagittal T1 images: [x]\n- Standard sagittal T2 images: [y]\n- Standard sagittal T2 SPACE images: [z]\n\nAn additional hidden test set (not available through Hugging Face) is available on the SPIDER Grand Challenge.", "## Image Resolution\n\nStandard sagittal T1 and T2 image resolution ranges from 3.3 x 0.33 x 0.33 mm to 4.8 x 0.90 x 0.90 mm. \nSagittal T2 SPACE sequence images had a near isotropic spatial resolution with a voxel size of 0.90 x 0.47 x 0.47 mm.\nSource", "## Dataset Curation\n\nThe data have been curated to enable users to load any of the following:\n\n- Raw image files\n- Raw segmentation masks\n- Numeric representations of images in tensor format\n- Numeric representations of segmentation masks in tensor format\n- Linked patient characteristics (limited to sex and age, if available)\n- Linked scanner characteristics", "### Source Data", "### Processing Steps\n\n(Specifics to be determined, but will include:)\n\n1. Conversion of .mha files to numeric representations\n2. Linking of segmentation mask numeric representations to image files\n3. Linking of patient and scanner characteristics to image files\n4. Cleaning of patient and scanner characteristics", "## Additional Information", "### License\n\nThe dataset is published under a CC-BY 4.0 license: URL\n\nJasper W. van der Graaf, Miranda L. van Hooff, Constantinus F. M. Buckens, Matthieu Rutten, \nJob L. C. van Susante, Robert Jan Kroeze, Marinus de Kleuver, Bram van Ginneken, Nikolas Lessmann. (2023).\n*Lumbar spine segmentation in MR images: a dataset and a public benchmark.* URL" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"medqa_train_instruction_format\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"medqa_train_instruction_format\"\n\nMore Information needed" ]

[ "task_categories:conversational", "size_categories:100K<n<1M", "language:ja", "region:us" ]

[]

[ "ja" ]

[ "# Abstruct\nThis is a multi-turn conversation dataset generated from the Japanese Wikipedia dataset using Orion14B-Chat. Commercial use is possible, but the license is complicated, so please read it carefully before using it.\nI generated V100x4 on 200 machines in about half a week.", "# License\n【Orion-14B Series】 Models Community License Agreement\nURL", "# Computing\nABCI\nURL" ]

[ "TAGS\n#task_categories-conversational #size_categories-100K<n<1M #language-Japanese #region-us \n", "# Abstruct\nThis is a multi-turn conversation dataset generated from the Japanese Wikipedia dataset using Orion14B-Chat. Commercial use is possible, but the license is complicated, so please read it carefully before using it.\nI generated V100x4 on 200 machines in about half a week.", "# License\n【Orion-14B Series】 Models Community License Agreement\nURL", "# Computing\nABCI\nURL" ]

[ "task_categories:question-answering", "task_categories:feature-extraction", "size_categories:1K<n<10K", "language:en", "license:odc-by", "arxiv:2401.15391", "region:us" ]

[ "2401.15391" ]

[ "en" ]

[ "# Dataset Card for Dataset Name\n\nA Dataset for Evaluating Retrieval-Augmented Generation Across Documents", "### Dataset Description\n\nMultiHop-RAG: a QA dataset to evaluate retrieval and reasoning across documents with metadata in the RAG pipelines. It contains 2556 queries, with evidence for each query distributed across 2 to 4 documents. The queries also involve document metadata, reflecting complex scenarios commonly found in real-world RAG applications.", "### Dataset Sources \n\n\n\n- Github: MultiHop-RAG\n- Paper: MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries\n\nBibTeX:" ]

[ "TAGS\n#task_categories-question-answering #task_categories-feature-extraction #size_categories-1K<n<10K #language-English #license-odc-by #arxiv-2401.15391 #region-us \n", "# Dataset Card for Dataset Name\n\nA Dataset for Evaluating Retrieval-Augmented Generation Across Documents", "### Dataset Description\n\nMultiHop-RAG: a QA dataset to evaluate retrieval and reasoning across documents with metadata in the RAG pipelines. It contains 2556 queries, with evidence for each query distributed across 2 to 4 documents. The queries also involve document metadata, reflecting complex scenarios commonly found in real-world RAG applications.", "### Dataset Sources \n\n\n\n- Github: MultiHop-RAG\n- Paper: MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries\n\nBibTeX:" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"coref_100\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"coref_100\"\n\nMore Information needed" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"JigsawTest\"", "## Source\n - TARA", "## Task\n - Determine which one of image2 (A) and image3 (B) can complete image1 (the lower right corner of image1 is missing).", "## Prompt\nConcise version:\n\n\nVerbose version:" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"JigsawTest\"", "## Source\n - TARA", "## Task\n - Determine which one of image2 (A) and image3 (B) can complete image1 (the lower right corner of image1 is missing).", "## Prompt\nConcise version:\n\n\nVerbose version:" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"Jigsaw\"", "## Source\n - TARA", "## Task\n - Determine which one of image2 (A) and image3 (B) can complete image1 (the lower right corner of image1 is missing)." ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"Jigsaw\"", "## Source\n - TARA", "## Task\n - Determine which one of image2 (A) and image3 (B) can complete image1 (the lower right corner of image1 is missing)." ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"counterfactual-babylm-pipps_and_keys_to_it_all_removal\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"counterfactual-babylm-pipps_and_keys_to_it_all_removal\"\n\nMore Information needed" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"hf-codegen-v2\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"hf-codegen-v2\"\n\nMore Information needed" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"hug_stack\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"hug_stack\"\n\nMore Information needed" ]

[ "task_categories:automatic-speech-recognition", "language:ms", "license:mit", "region:us" ]

[]

[ "ms" ]

[ "# Sebut Perkataan\n\n1. 'sebut-perkataan-man' voice by Husein Zolkepli\n2. 'tolong-sebut' voice by Khalil Nooh\n3. 'sebut-perkataan-woman' voice by Mas Aisyah Ahmad\n4. Recorded using low-end tech microphones." ]

[ "TAGS\n#task_categories-automatic-speech-recognition #language-Malay (macrolanguage) #license-mit #region-us \n", "# Sebut Perkataan\n\n1. 'sebut-perkataan-man' voice by Husein Zolkepli\n2. 'tolong-sebut' voice by Khalil Nooh\n3. 'sebut-perkataan-woman' voice by Mas Aisyah Ahmad\n4. Recorded using low-end tech microphones." ]

[ "task_categories:text-generation", "task_categories:question-answering", "task_categories:conversational", "language:ru", "license:apache-2.0", "region:us" ]

[]

[ "ru" ]

[]

[ "TAGS\n#task_categories-text-generation #task_categories-question-answering #task_categories-conversational #language-Russian #license-apache-2.0 #region-us \n" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"voxceleb1_unit\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"voxceleb1_unit\"\n\nMore Information needed" ]

[ "distilabel", "synthetic", "region:us" ]

[]

[]

[ "# Dataset Card for \"dpo-mix-7k\"\n\nA cocktail with our latest DPO datasets: orca pairs, ultrafeedback ratings, and capybara-dpo" ]

[ "TAGS\n#distilabel #synthetic #region-us \n", "# Dataset Card for \"dpo-mix-7k\"\n\nA cocktail with our latest DPO datasets: orca pairs, ultrafeedback ratings, and capybara-dpo" ]

[ "license:mit", "region:us" ]

[]

[]

[]

[ "TAGS\n#license-mit #region-us \n" ]

[ "region:us" ]

[]

[]

[]

[ "TAGS\n#region-us \n" ]

[ "task_categories:text-classification", "task_ids:multi-class-classification", "annotations_creators:crowdsourced", "language_creators:crowdsourced", "size_categories:n<2K", "source_datasets:reddit", "language:en", "license:cc", "natural-language-understanding", "ideology classification", "text classification", "region:us" ]

[]

[ "en" ]

[ "# Dataset Card for Dataset Name\n\n\n\nThis dataset aims to be a tool to help trace linguistic patterns in the reddit posts from members who partake in the internet centric pill ideologies, known as blackpill, red pill, blue pill.", "## Dataset Details", "### Dataset Description\n\nA few of the major groups' posts have been coalesced into one dataset, all from different years. There are more than 200 posts per the major pill groups on reddit (red pill rebooted, blue pill, black pill, married red pill, red pill women, and feminism as a counterpoint of reference). The group of feminism was added as a juxtaposition against red pill women, in oder to allow researchers to explore those dichotomies. For researchers, the value will be in identifying or classifying the types of words that make one ideology more prominent than the other.\n\n- Curated by: [steamcyclone]\n- Funded by [optional]: \n- Shared by [optional]: [steamcyclone]\n- Language(s) (NLP): [EN]\n- License: [CC]", "### Dataset Sources [optional]\n\n\n\n- Repository: [This is the only source]", "## Uses\n\nThe main usage of this dataset is to study linguistic patterns. Running models and detecting word usage per groups, as well as overlaps across groups is an ideal use for this dataset. With the rise of the loneliness epidemic, any insights that come from this are welcome.", "### Direct Use\n\nThe suitable use cases are to multi-class classification, word clustering or semantic clustering per different groups, summarization modeling, text parsing, and any other natural language processing task.", "### Out-of-Scope Use\n\nThis dataset is not meant to be utilized to demonize or mock certain online communities for the trials in life in which individuals find themselves. If the viewer's agenda is to push forward some misandrist or misogynistic agenda, please ignore this dataset.", "## Dataset Structure\n\n\n\nCurrently, this dataset contains \n\n- subreddit of the post : string,\n- postid : string\n- title of the post: string\n- text of the post (where applicable) : string\n- url (if something was embedded) : string\\\n- score : int32\n- author : string\n- date : int64", "## Dataset Creation", "### Curation Rationale\n\nWith the rise of the loneliness epidemic and the radicalization of internet content pitting men and women against each other, it is important to seek understanding of the root of the problem. Depending on whom you ask, you'll get a plethora of answers. Jordan Peterson describes it as some type of post-modernist feminist liberalism problem. The Andrew Tates and other conservative archetypes blame the loss of traditionalism. Others blame dating apps and its selection bias effects. Within each of the major pill ideologies, with the exception of the BlackPill, men blame women, and women blame men. \n\nUnfortunately, male spaces, as substantiated by research and media coverage, in recent years have only been able to exist on the internet, and counter-spaces have emerged to challenge the views held in the differing ideologies.\n\nIn short, according to archetypical definitions\n- the red pill is the emancipation of the masculinity in a feminized age and understanding mating strategies with women. \n- the blue pill is the satire of the red pill, often run by women.\n- the black pill is meant to bridge the gaps across the red, pink, and blue pills in order to land on a ground truth.\n- the pink pill is about improving the female image by augmenting sexual marketplace value.", "### Source Data\n\nEach record contains a reddit post, approximately 200 per group, and has a key title and a post with words to display the intended message by the author.", "#### Data Collection and Processing\n\n\n\nIn progress. \n\nHowever, the plan is to increase the amount of records and leverage the ChatGpt API to summarize the messages into categories. In addition, the dates have to be cleaned a little, in order to add use for researches. I am also not sure if I can retrieve comments per post, further augmenting the data.", "#### Who are the source data producers?\n\nThe producers of the data are the various redditors who have participated in these spaces.", "### Annotations [optional]\n\nAn annotation that is not part of the collection will be the ChatGPT summarizations (future). The subreddit labels are merely the origins of the posts.", "#### Annotation process\n\n\n\nThe origin of the posts are the labels of the records.", "#### Who are the annotators?\n\nI and the subreddit origin are the label annotators.", "#### Personal and Sensitive Information\n\n\n\nThis dataset contains no personally identifiable information with the exception of embedded youtube links. Those links may lead to videos where the impact of the content is unknown.", "## Bias, Risks, and Limitations\n\n\n\nA major caveat is that the pink pill and original red pill groups are shadow banned, impeding their scraping process. This is a flaw I recognize because the original red pill movement, which started in books by authors, propagated itself through its internet (reddit) variant, and it spawned all the other pills.\n\nAnother bias point is that there is more red pill content, as a means to compensate for the ban of the original red pill subreddit. \n\nAs such, I caution researchers to balance their datasets where necessary.", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. Remember that this dataset is not a tool for reckless and hateful political agendas.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]\n\nPill ideologies :\n\nIn short, according to archetypical definitions\n- the red pill is the emancipation of the masculinity in a feminized age and understanding mating strategies with women. \n- the blue pill is the satire of the red pill, often run by women.\n- the black pill is meant to bridge the gaps across the red, pink, and blue pills in order to land on a ground truth.\n- the pink pill is about improving the female image by augmenting sexual marketplace value.", "## Dataset Card Authors [optional]\n\nsteamcyclone, all the redditors from the subreddits in the authors columns.", "## Dataset Card Contact\n\n- N/A" ]

[ "TAGS\n#task_categories-text-classification #task_ids-multi-class-classification #annotations_creators-crowdsourced #language_creators-crowdsourced #size_categories-n<2K #source_datasets-reddit #language-English #license-cc #natural-language-understanding #ideology classification #text classification #region-us \n", "# Dataset Card for Dataset Name\n\n\n\nThis dataset aims to be a tool to help trace linguistic patterns in the reddit posts from members who partake in the internet centric pill ideologies, known as blackpill, red pill, blue pill.", "## Dataset Details", "### Dataset Description\n\nA few of the major groups' posts have been coalesced into one dataset, all from different years. There are more than 200 posts per the major pill groups on reddit (red pill rebooted, blue pill, black pill, married red pill, red pill women, and feminism as a counterpoint of reference). The group of feminism was added as a juxtaposition against red pill women, in oder to allow researchers to explore those dichotomies. For researchers, the value will be in identifying or classifying the types of words that make one ideology more prominent than the other.\n\n- Curated by: [steamcyclone]\n- Funded by [optional]: \n- Shared by [optional]: [steamcyclone]\n- Language(s) (NLP): [EN]\n- License: [CC]", "### Dataset Sources [optional]\n\n\n\n- Repository: [This is the only source]", "## Uses\n\nThe main usage of this dataset is to study linguistic patterns. Running models and detecting word usage per groups, as well as overlaps across groups is an ideal use for this dataset. With the rise of the loneliness epidemic, any insights that come from this are welcome.", "### Direct Use\n\nThe suitable use cases are to multi-class classification, word clustering or semantic clustering per different groups, summarization modeling, text parsing, and any other natural language processing task.", "### Out-of-Scope Use\n\nThis dataset is not meant to be utilized to demonize or mock certain online communities for the trials in life in which individuals find themselves. If the viewer's agenda is to push forward some misandrist or misogynistic agenda, please ignore this dataset.", "## Dataset Structure\n\n\n\nCurrently, this dataset contains \n\n- subreddit of the post : string,\n- postid : string\n- title of the post: string\n- text of the post (where applicable) : string\n- url (if something was embedded) : string\\\n- score : int32\n- author : string\n- date : int64", "## Dataset Creation", "### Curation Rationale\n\nWith the rise of the loneliness epidemic and the radicalization of internet content pitting men and women against each other, it is important to seek understanding of the root of the problem. Depending on whom you ask, you'll get a plethora of answers. Jordan Peterson describes it as some type of post-modernist feminist liberalism problem. The Andrew Tates and other conservative archetypes blame the loss of traditionalism. Others blame dating apps and its selection bias effects. Within each of the major pill ideologies, with the exception of the BlackPill, men blame women, and women blame men. \n\nUnfortunately, male spaces, as substantiated by research and media coverage, in recent years have only been able to exist on the internet, and counter-spaces have emerged to challenge the views held in the differing ideologies.\n\nIn short, according to archetypical definitions\n- the red pill is the emancipation of the masculinity in a feminized age and understanding mating strategies with women. \n- the blue pill is the satire of the red pill, often run by women.\n- the black pill is meant to bridge the gaps across the red, pink, and blue pills in order to land on a ground truth.\n- the pink pill is about improving the female image by augmenting sexual marketplace value.", "### Source Data\n\nEach record contains a reddit post, approximately 200 per group, and has a key title and a post with words to display the intended message by the author.", "#### Data Collection and Processing\n\n\n\nIn progress. \n\nHowever, the plan is to increase the amount of records and leverage the ChatGpt API to summarize the messages into categories. In addition, the dates have to be cleaned a little, in order to add use for researches. I am also not sure if I can retrieve comments per post, further augmenting the data.", "#### Who are the source data producers?\n\nThe producers of the data are the various redditors who have participated in these spaces.", "### Annotations [optional]\n\nAn annotation that is not part of the collection will be the ChatGPT summarizations (future). The subreddit labels are merely the origins of the posts.", "#### Annotation process\n\n\n\nThe origin of the posts are the labels of the records.", "#### Who are the annotators?\n\nI and the subreddit origin are the label annotators.", "#### Personal and Sensitive Information\n\n\n\nThis dataset contains no personally identifiable information with the exception of embedded youtube links. Those links may lead to videos where the impact of the content is unknown.", "## Bias, Risks, and Limitations\n\n\n\nA major caveat is that the pink pill and original red pill groups are shadow banned, impeding their scraping process. This is a flaw I recognize because the original red pill movement, which started in books by authors, propagated itself through its internet (reddit) variant, and it spawned all the other pills.\n\nAnother bias point is that there is more red pill content, as a means to compensate for the ban of the original red pill subreddit. \n\nAs such, I caution researchers to balance their datasets where necessary.", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. Remember that this dataset is not a tool for reckless and hateful political agendas.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]\n\nPill ideologies :\n\nIn short, according to archetypical definitions\n- the red pill is the emancipation of the masculinity in a feminized age and understanding mating strategies with women. \n- the blue pill is the satire of the red pill, often run by women.\n- the black pill is meant to bridge the gaps across the red, pink, and blue pills in order to land on a ground truth.\n- the pink pill is about improving the female image by augmenting sexual marketplace value.", "## Dataset Card Authors [optional]\n\nsteamcyclone, all the redditors from the subreddits in the authors columns.", "## Dataset Card Contact\n\n- N/A" ]

[ "task_categories:text-generation", "language:de", "region:us" ]

[]

[ "de" ]

[]

[ "TAGS\n#task_categories-text-generation #language-German #region-us \n" ]

[ "task_categories:text-generation", "language:de", "region:us" ]

[]

[ "de" ]

[]

[ "TAGS\n#task_categories-text-generation #language-German #region-us \n" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"ultrafeedback_binarized_narrow\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"ultrafeedback_binarized_narrow\"\n\nMore Information needed" ]

[ "task_categories:text-generation", "language:de", "region:us" ]

[]

[ "de" ]

[]

[ "TAGS\n#task_categories-text-generation #language-German #region-us \n" ]

[ "task_categories:text-generation", "language:de", "region:us" ]

[]

[ "de" ]

[]

[ "TAGS\n#task_categories-text-generation #language-German #region-us \n" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for Dataset Name\n\n\n\nThis dataset card aims to be a base template for new datasets. It has been generated using this raw template.", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "TAGS\n#region-us \n", "# Dataset Card for Dataset Name\n\n\n\nThis dataset card aims to be a base template for new datasets. It has been generated using this raw template.", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "region:us" ]

[]

[]

[ "# AI Generated Image for Image Classification\n\n\n\nThis dataset contains AI generated images by Midjourney and Human images taken from Imagenet. The dataset is meant for Image Classification tasks.", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: Deepankar Sharma" ]

[ "TAGS\n#region-us \n", "# AI Generated Image for Image Classification\n\n\n\nThis dataset contains AI generated images by Midjourney and Human images taken from Imagenet. The dataset is meant for Image Classification tasks.", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: Deepankar Sharma" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"maywell_ko_wikidata_QA_12800\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"maywell_ko_wikidata_QA_12800\"\n\nMore Information needed" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"keyword_ner\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"keyword_ner\"\n\nMore Information needed" ]

[ "language:ar", "region:us" ]

[]

[ "ar" ]

[ "# Dataset Card for Dataset Name\n\n\n\nThis dataset card aims to be a base template for new datasets. It has been generated using this raw template.", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "TAGS\n#language-Arabic #region-us \n", "# Dataset Card for Dataset Name\n\n\n\nThis dataset card aims to be a base template for new datasets. It has been generated using this raw template.", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "region:us" ]

[]

[]

[ "## Formatting\n\nFormatting is compliant with ChatML. \"input\" is the context and \"output\" is the expected model output to train on.", "## Details\n\nSee the repo file 'generate_dataset.py' for exactly this dataset was generated.\n\nAn opinionated and filtered mix of the following datasets:\n\n- argilla/ultrafeedback-binarized-preferences-cleaned\n- heegyu/glaive-function-calling-v2-formatted\n- berkeley-nest/Nectar\n- argilla/distilabel-math-preference-dpo", "## argilla/ultrafeedback-binarized-preferences-cleaned\n\nFilter: \n- 'chosen-rating == 5' AND\n- 'len(chosen) == 2'", "## heegyu/glaive-function-calling-v2-formatted\n\nFilter: \n- 'function_description != '''\n\nTransforms: \n- Added a system message randomly selected from a pool of generic system messages.", "## berkeley-nest/Nectar\n\nFilter: \n- has an answer with 'rank == 1' AND\n- 'turns > 1' AND\n- 'good_natured == True' AND\n- answer.to_lower() does not start with \"i'm sorry\"", "## argilla/distilabel-math-preference-dpo\n\nFilter: \n- 'chosen-rating >= 9'\n\nTransforms: \n- Added a system message randomly selected from a pool of generic system messages.\n- Removed the phrase \"Take a deep breath, think step by step, and give an accurate response\"", "## Global formatting\n\nAll the above datasets were formatted to comply with ChatML." ]

[ "TAGS\n#region-us \n", "## Formatting\n\nFormatting is compliant with ChatML. \"input\" is the context and \"output\" is the expected model output to train on.", "## Details\n\nSee the repo file 'generate_dataset.py' for exactly this dataset was generated.\n\nAn opinionated and filtered mix of the following datasets:\n\n- argilla/ultrafeedback-binarized-preferences-cleaned\n- heegyu/glaive-function-calling-v2-formatted\n- berkeley-nest/Nectar\n- argilla/distilabel-math-preference-dpo", "## argilla/ultrafeedback-binarized-preferences-cleaned\n\nFilter: \n- 'chosen-rating == 5' AND\n- 'len(chosen) == 2'", "## heegyu/glaive-function-calling-v2-formatted\n\nFilter: \n- 'function_description != '''\n\nTransforms: \n- Added a system message randomly selected from a pool of generic system messages.", "## berkeley-nest/Nectar\n\nFilter: \n- has an answer with 'rank == 1' AND\n- 'turns > 1' AND\n- 'good_natured == True' AND\n- answer.to_lower() does not start with \"i'm sorry\"", "## argilla/distilabel-math-preference-dpo\n\nFilter: \n- 'chosen-rating >= 9'\n\nTransforms: \n- Added a system message randomly selected from a pool of generic system messages.\n- Removed the phrase \"Take a deep breath, think step by step, and give an accurate response\"", "## Global formatting\n\nAll the above datasets were formatted to comply with ChatML." ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"orca-tr\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"orca-tr\"\n\nMore Information needed" ]

[ "vdf", "vector-io", "vector-dataset", "vector-embeddings", "region:us" ]

[]

[]

[]

[ "TAGS\n#vdf #vector-io #vector-dataset #vector-embeddings #region-us \n" ]

[ "vdf", "vector-io", "vector-dataset", "vector-embeddings", "region:us" ]

[]

[]

[]

[ "TAGS\n#vdf #vector-io #vector-dataset #vector-embeddings #region-us \n" ]

[ "vdf", "vector-io", "vector-dataset", "vector-embeddings", "region:us" ]

[]

[]

[]

[ "TAGS\n#vdf #vector-io #vector-dataset #vector-embeddings #region-us \n" ]

[ "arxiv:2310.15051", "region:us" ]

[ "2310.15051" ]

[]

[ "# TeleQnA", "## Introduction\n\nTeleQnA is a comprehensive dataset tailored to assess the knowledge of Large Language Models (LLMs) in the field of telecommunications. It encompasses 10,000 multiple-choice questions distributed across five distinct categories:\n\n- Lexicon: This category comprises 500 questions that delve into the realm of general telecom terminology and definitions.\n\n- Research overview: Comprising 2,000 questions, this category provides a broad overview of telecom research, spanning a wide spectrum of telecom-related topics.\n\n- Research publications: With 4,500 questions, this category contains detailed inquiries regarding multi-disciplanary research in telecommunications, drawing from a variety of sources such as transactions and conferences proceedings.\n\n- Standards overview: This category consists of 1,000 questions related to summaries of standards from multiple standarization bodies like 3GPP and IEEE.\n\n- Standards specifications: With 2,000 questions, this category explores the technical specifications and practical implementations of telecommunications systems, leveraging information from standardization bodies like 3GPP and IEEE.\n\nFor more in-depth information about the dataset and the generation process, please refer to our paper by following this link. To prevent inadvertent data contamination with models trained using GitHub data, we have implemented a password protection measure for unzipping the dataset. The password is .", "## Dataset Format\n\nEach question is represented in JSON format, comprising five distinct fields:\n\n- Question: This field consists of a string that presents the question associated with a specific concept within the telecommunications domain.\n\n- Options: This field comprises a set of strings representing the various answer options.\n\n- Answer: This field contains a string that adheres to the format ’option ID: Answer’ and presents the correct response to the question. A single option is correct; however, options may include choices like “All of the Above” or “Both options 1 and 2”.\n\n- Explanation: This field encompasses a string that clarifies the reasoning behind the correct answer.\n\n- Category: This field includes a label identifying the source category (e.g., lexicon, research overview, etc.).", "## Dataset Instance \n\nAn example of the dataset is provided below:", "## Experiments Code\n\nThe provided code allows to evaluate the performance of OpenAI's models (e.g., GPT-3.5). To do so, follow the below steps:\n\n- Clone the repository\n- Unzip URL using the password \n- Install the required dependencies using the following command:\n\n\n\n- Insert OpenAI's API key into the evaluation_tools script.\n- Run the command below\n\n \n\nUpon completion, a .txt file in JSON format is generated. This file contains the original dataset, with two additional fields added to each question:\n\n- tested answer: This field contains the answer chosen by the tested model.\n\n- correct: This field is marked as \"True\" when the tested answer matches the designated correct answer in the dataset.\n\nIf you would like to use the data or code, please cite the paper:" ]

[ "TAGS\n#arxiv-2310.15051 #region-us \n", "# TeleQnA", "## Introduction\n\nTeleQnA is a comprehensive dataset tailored to assess the knowledge of Large Language Models (LLMs) in the field of telecommunications. It encompasses 10,000 multiple-choice questions distributed across five distinct categories:\n\n- Lexicon: This category comprises 500 questions that delve into the realm of general telecom terminology and definitions.\n\n- Research overview: Comprising 2,000 questions, this category provides a broad overview of telecom research, spanning a wide spectrum of telecom-related topics.\n\n- Research publications: With 4,500 questions, this category contains detailed inquiries regarding multi-disciplanary research in telecommunications, drawing from a variety of sources such as transactions and conferences proceedings.\n\n- Standards overview: This category consists of 1,000 questions related to summaries of standards from multiple standarization bodies like 3GPP and IEEE.\n\n- Standards specifications: With 2,000 questions, this category explores the technical specifications and practical implementations of telecommunications systems, leveraging information from standardization bodies like 3GPP and IEEE.\n\nFor more in-depth information about the dataset and the generation process, please refer to our paper by following this link. To prevent inadvertent data contamination with models trained using GitHub data, we have implemented a password protection measure for unzipping the dataset. The password is .", "## Dataset Format\n\nEach question is represented in JSON format, comprising five distinct fields:\n\n- Question: This field consists of a string that presents the question associated with a specific concept within the telecommunications domain.\n\n- Options: This field comprises a set of strings representing the various answer options.\n\n- Answer: This field contains a string that adheres to the format ’option ID: Answer’ and presents the correct response to the question. A single option is correct; however, options may include choices like “All of the Above” or “Both options 1 and 2”.\n\n- Explanation: This field encompasses a string that clarifies the reasoning behind the correct answer.\n\n- Category: This field includes a label identifying the source category (e.g., lexicon, research overview, etc.).", "## Dataset Instance \n\nAn example of the dataset is provided below:", "## Experiments Code\n\nThe provided code allows to evaluate the performance of OpenAI's models (e.g., GPT-3.5). To do so, follow the below steps:\n\n- Clone the repository\n- Unzip URL using the password \n- Install the required dependencies using the following command:\n\n\n\n- Insert OpenAI's API key into the evaluation_tools script.\n- Run the command below\n\n \n\nUpon completion, a .txt file in JSON format is generated. This file contains the original dataset, with two additional fields added to each question:\n\n- tested answer: This field contains the answer chosen by the tested model.\n\n- correct: This field is marked as \"True\" when the tested answer matches the designated correct answer in the dataset.\n\nIf you would like to use the data or code, please cite the paper:" ]

[ "license:cc-by-nc-4.0", "region:us" ]

[]

[]

[ "# Dataset Card for \"Promoter Validation Data\"", "## Table of Contents\n\n- Overview\n- Dataset Description\n - Data Collection and Processing\n - Dataset Structure\n- Data Partitioning and Utilization\n- Dataset Splits\n- Dataset Fields\n- Use Cases & Limitations\n- How to Use\n- Citation\n- Contact", "## Overview\n\nThis dataset contains promoter sequences used for validation purposes in genetic research, focusing on prokaryotic promoters. It serves to facilitate the study of gene expression regulation, providing a comprehensive set of promoter sequences from various organisms.\n\nThe dataset comprises known promoter sequences from 75 organisms, retrieved from the Prokaryotic Promoter Database (PPD), alongside non-promoter sequences obtained from the NCBI RefSeq database (specifically sampled from CDS regions). It also includes non-promoter sequences constructed via higher and zero-order Markov chains that mirror compositional characteristics of known promoters. Additionally, an independent test set, focusing on *E. coli* sigma70 promoters, was employed, curated by Cassiano and Silva-Rocha (2020).\n\nA balanced distribution approach was adopted to even out the number of positive and negative samples, with the dataset being systematically divided into training, validation, and test subsets. This stratification underpins a thorough evaluation of the model efficacy.\n\n\n!Promoter Dataset Overview\n*Figure 1: Promoter dataset - overview*\n\nThe prokaryotic promoter sequences are typically 81bp long, ensuring compatibility with most tools' input prerequisites, particularly around the putative TSS region interval $[-60, +20]$. Our positive dataset encompasses promoter sequences from various species, predominantly found on both chromosomes and plasmids. Promoters included in the independent test set, based on exact match, were excluded from the training data. Species and contigs were mapped to NCBI assembly and sequence accessions.\n\nTo curate comprehensive non-promoter sequences (negative samples), we employed three strategies:\n1. Using non-promoter sequences (CDS -- Coding Sequences).\n2. Random sequences generated with a 3rd-order Markov chain.\n3. Pure random sequences (0-order Markov chain) as proposed by Cassiano and Silva-Rocha (2020).\n\nThe distribution of this composite dataset was 40% CDS, 40% Markov-derived random sequences, and 20% pure random sequences (0-order Markov chain). One practical application of promoter detection in coding sequences is to check whether an unintentional promoter is injected or can be located inside a modified or designed coding sequence region, causing disruption. To cover this use-case, we incorporated the coding regions into our training and evaluation dataset. The CDS sequences were extracted from the genomic sequences of contigs, based on annotations from NCBI. The 81bp long CDS region samples were selected based on the NCBI-provided annotations for the available contigs with respect to the underlying species. The promoter regions often contain AT-rich sequences, i.e., TATA box. To capture and model the AT-rich regions, we applied 3rd and 0 order Markov chains to generate sequence examples that reflect the compositional property of known promoters.", "## Dataset Description", "### Data Collection and Processing\n\n- Data Source: The positive samples, known promoters, are primarily drawn from the Prokaryotic Promoter Database (PPD), containing experimentally validated promoter sequences from 75 organisms. Non-promoter sequences are obtained from the NCBI RefSeq database, sampled specifically from CDS regions.\n- Preprocessing: The dataset includes non-promoter sequences constructed via higher and zero-order Markov chains, which mirror compositional characteristics of known promoters. An independent test set based on _E.coli_ sigma70 promoters is also included.", "### Dataset Structure\n\n- Dataset Splits: The dataset is systematically divided into training, validation, and test subsets.\n- Data Fields:\n - 'segment_id': Unique identifier for each segment.\n - 'ppd_original_SpeciesName': Original species name from the PPD.\n - 'Strand': The strand of the DNA sequence.\n - 'segment': The DNA sequence of the promoter region.\n - 'label': The label indicating whether the sequence is a promoter or non-promoter.\n - 'L': Length of the DNA sequence.\n - 'prom_class': The class of the promoter.\n - 'y': Binary label indicating the presence of a promoter.", "## Data Partitioning and Utilization\n\nTo ensure comprehensive evaluation, the dataset was split randomly into three parts:\n1. Training set: 80% of the total data for model development and training.\n2. Validation set: 10% of the data, aiding in fine-tuning model parameters.\n3. Test set: Remaining 10% of the data, crucial for unbiased model performance evaluation.", "## Dataset Splits\n\n- Training Set: Primary dataset used for model training.\n- Test Set (Sigma70): Independent test set focusing on _E.coli_ sigma70 promoters.\n- Multispecies Set: Additional test set including various species, ensuring generalization across different organisms.", "## Use Cases & Limitations\n\nThis dataset is intended for use in training models for promoter prediction and understanding gene expression regulation. Limitations include the specific focus on prokaryotic promoters and the method of sequence generation using Markov chains.", "## How to Use\n\nYou can load the dataset using the Hugging Face 'datasets' library:", "## Contact Information\n\nFor any questions, feedback, or contributions regarding the datasets or ProkBERT, please feel free to reach out:\n\n- Name: Balázs Ligeti\n- Email: obalasz@URL\n\nWe welcome your input and collaboration to improve our resources and research.\n\n\n\n'''bibtex\n@Article{ProkBERT2024,\n author = {Ligeti, Balázs et al.},\n journal = {Frontiers in Microbiology},\n title = {{ProkBERT} family: genomic language models},\n year = {2024},\n volume = {14},\n URL = {URL\n DOI = {10.3389/fmicb.2023.1331233}\n}" ]

[ "TAGS\n#license-cc-by-nc-4.0 #region-us \n", "# Dataset Card for \"Promoter Validation Data\"", "## Table of Contents\n\n- Overview\n- Dataset Description\n - Data Collection and Processing\n - Dataset Structure\n- Data Partitioning and Utilization\n- Dataset Splits\n- Dataset Fields\n- Use Cases & Limitations\n- How to Use\n- Citation\n- Contact", "## Overview\n\nThis dataset contains promoter sequences used for validation purposes in genetic research, focusing on prokaryotic promoters. It serves to facilitate the study of gene expression regulation, providing a comprehensive set of promoter sequences from various organisms.\n\nThe dataset comprises known promoter sequences from 75 organisms, retrieved from the Prokaryotic Promoter Database (PPD), alongside non-promoter sequences obtained from the NCBI RefSeq database (specifically sampled from CDS regions). It also includes non-promoter sequences constructed via higher and zero-order Markov chains that mirror compositional characteristics of known promoters. Additionally, an independent test set, focusing on *E. coli* sigma70 promoters, was employed, curated by Cassiano and Silva-Rocha (2020).\n\nA balanced distribution approach was adopted to even out the number of positive and negative samples, with the dataset being systematically divided into training, validation, and test subsets. This stratification underpins a thorough evaluation of the model efficacy.\n\n\n!Promoter Dataset Overview\n*Figure 1: Promoter dataset - overview*\n\nThe prokaryotic promoter sequences are typically 81bp long, ensuring compatibility with most tools' input prerequisites, particularly around the putative TSS region interval $[-60, +20]$. Our positive dataset encompasses promoter sequences from various species, predominantly found on both chromosomes and plasmids. Promoters included in the independent test set, based on exact match, were excluded from the training data. Species and contigs were mapped to NCBI assembly and sequence accessions.\n\nTo curate comprehensive non-promoter sequences (negative samples), we employed three strategies:\n1. Using non-promoter sequences (CDS -- Coding Sequences).\n2. Random sequences generated with a 3rd-order Markov chain.\n3. Pure random sequences (0-order Markov chain) as proposed by Cassiano and Silva-Rocha (2020).\n\nThe distribution of this composite dataset was 40% CDS, 40% Markov-derived random sequences, and 20% pure random sequences (0-order Markov chain). One practical application of promoter detection in coding sequences is to check whether an unintentional promoter is injected or can be located inside a modified or designed coding sequence region, causing disruption. To cover this use-case, we incorporated the coding regions into our training and evaluation dataset. The CDS sequences were extracted from the genomic sequences of contigs, based on annotations from NCBI. The 81bp long CDS region samples were selected based on the NCBI-provided annotations for the available contigs with respect to the underlying species. The promoter regions often contain AT-rich sequences, i.e., TATA box. To capture and model the AT-rich regions, we applied 3rd and 0 order Markov chains to generate sequence examples that reflect the compositional property of known promoters.", "## Dataset Description", "### Data Collection and Processing\n\n- Data Source: The positive samples, known promoters, are primarily drawn from the Prokaryotic Promoter Database (PPD), containing experimentally validated promoter sequences from 75 organisms. Non-promoter sequences are obtained from the NCBI RefSeq database, sampled specifically from CDS regions.\n- Preprocessing: The dataset includes non-promoter sequences constructed via higher and zero-order Markov chains, which mirror compositional characteristics of known promoters. An independent test set based on _E.coli_ sigma70 promoters is also included.", "### Dataset Structure\n\n- Dataset Splits: The dataset is systematically divided into training, validation, and test subsets.\n- Data Fields:\n - 'segment_id': Unique identifier for each segment.\n - 'ppd_original_SpeciesName': Original species name from the PPD.\n - 'Strand': The strand of the DNA sequence.\n - 'segment': The DNA sequence of the promoter region.\n - 'label': The label indicating whether the sequence is a promoter or non-promoter.\n - 'L': Length of the DNA sequence.\n - 'prom_class': The class of the promoter.\n - 'y': Binary label indicating the presence of a promoter.", "## Data Partitioning and Utilization\n\nTo ensure comprehensive evaluation, the dataset was split randomly into three parts:\n1. Training set: 80% of the total data for model development and training.\n2. Validation set: 10% of the data, aiding in fine-tuning model parameters.\n3. Test set: Remaining 10% of the data, crucial for unbiased model performance evaluation.", "## Dataset Splits\n\n- Training Set: Primary dataset used for model training.\n- Test Set (Sigma70): Independent test set focusing on _E.coli_ sigma70 promoters.\n- Multispecies Set: Additional test set including various species, ensuring generalization across different organisms.", "## Use Cases & Limitations\n\nThis dataset is intended for use in training models for promoter prediction and understanding gene expression regulation. Limitations include the specific focus on prokaryotic promoters and the method of sequence generation using Markov chains.", "## How to Use\n\nYou can load the dataset using the Hugging Face 'datasets' library:", "## Contact Information\n\nFor any questions, feedback, or contributions regarding the datasets or ProkBERT, please feel free to reach out:\n\n- Name: Balázs Ligeti\n- Email: obalasz@URL\n\nWe welcome your input and collaboration to improve our resources and research.\n\n\n\n'''bibtex\n@Article{ProkBERT2024,\n author = {Ligeti, Balázs et al.},\n journal = {Frontiers in Microbiology},\n title = {{ProkBERT} family: genomic language models},\n year = {2024},\n volume = {14},\n URL = {URL\n DOI = {10.3389/fmicb.2023.1331233}\n}" ]

[ "task_categories:question-answering", "language:en", "license:mit", "arxiv:2402.09910", "region:us" ]

[ "2402.09910" ]

[ "en" ]

[ "# arXivTection Dataset\nThe arXivTection dataset serves as a benchmark designed for the task of detecting pretraining data from Large Language models.\n\nThe dataset consists of 50 research papers extracted from arXiv. \n- 25 published in 2023: Non-Training data, \"_label_\" column = 0.\n- 25 published before 2022: Training data, \"_label_\" column = 1.\n\nFrom each paper ≈ 30 passages are extracted. Each passage is paraphrased 3 times using the Language Model Claude v2.0. <br>\nThe \"_Answer_\" column indicates which of the passages is the real excerpt.<br>\nPassages are extracted to be on average ≈ 128 tokens in length.\n\n<br>", "# Testing Models on arXivTection\nOur dataset is planned to be used on a Multiple-Choice-Question-Answering format. Nonetheless, it is compatible to be used with other pretraining data detection methods.<br>\nOur GitHub repository contains example scripts to evaluate models on our dataset.\n\n<br>", "# Compatibility\nThe Multiple-Choice-Question-Answering task with our Dataset is designed to be applied to various models, such as:<br>\n- LLaMA-2\n- Mistral\n- Mixtral\n- Chat-GPT (gpt-3.5-turbo-instruct)\n- GPT-3 (text-davinci-003)\n- Claude \n\n<br>", "# Loading the Dataset\n\n\n\n<br>", "# Citation\n\n\n\n<details>\n <summary> Research Papers Used</summary>\n [1] Attanasio, G., Plaza-del-Arco, F. M., Nozza, D., & Lauscher, A. (2023). A Tale of Pronouns: Interpretability Informs Gender Bias Mitigation for Fairer Instruction-Tuned Machine Translation. arXiv preprint arXiv:2310.12127. <br>\n [2] Shi, Y., Wu, L., & Shao, M. (2023). Adaptive End-to-End Metric Learning for Zero-Shot Cross-Domain Slot Filling. arXiv preprint arXiv:2310.15294. <br>\n [3] Keleg, A., Goldwater, S., & Magdy, W. (2023). ALDi: Quantifying the arabic level of dialectness of text. arXiv preprint arXiv:2310.13747. <br>\n [4] Su, Y., Ji, Y., Li, J., Ye, H., & Zhang, M. (2023, December). Beware of Model Collapse! Fast and Stable Test-time Adaptation for Robust Question Answering. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (pp. 12998-13011). <br>\n [5] Chang, Y., Lo, K., Goyal, T., & Iyyer, M. (2023). BooookScore: A systematic exploration of book-length summarization in the era of LLMs. arXiv preprint arXiv:2310.00785. <br>\n [6] Karamolegkou, A., Li, J., Zhou, L., & Søgaard, A. (2023). Copyright Violations and Large Language Models. arXiv preprint arXiv:2310.13771. <br>\n [7] Weissweiler, L., Hofmann, V., Kantharuban, A., Cai, A., Dutt, R., Hengle, A., ... & Mortensen, D. R. (2023). Counting the Bugs in ChatGPT's Wugs: A Multilingual Investigation into the Morphological Capabilities of a Large Language Model. arXiv preprint arXiv:2310.15113. <br>\n [8] Li, Z., & Zhang, Y. (2023, December). Cultural Concept Adaptation on Multimodal Reasoning. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (pp. 262-276). <br>\n [9] Jiang, W., Mao, Q., Li, J., Lin, C., Yang, W., Deng, T., & Wang, Z. (2023). DisCo: Distilled Student Models Co-training for Semi-supervised Text Mining. arXiv preprint arXiv:2305.12074. <br>\n [10] Zhu, Y., Si, J., Zhao, Y., Zhu, H., Zhou, D., & He, Y. (2023). EXPLAIN, EDIT, GENERATE: Rationale-Sensitive Counterfactual Data Augmentation for Multi-hop Fact Verification. arXiv preprint arXiv:2310.14508. <br>\n [11] Hada, R., Seth, A., Diddee, H., & Bali, K. (2023). ''Fifty Shades of Bias'': Normative Ratings of Gender Bias in GPT Generated English Text. arXiv preprint arXiv:2310.17428. <br>\n [12] Song, Y., & Dhariwal, P. (2023). Improved techniques for training consistency models. arXiv preprint arXiv:2310.14189. <br>\n [13] Xu, W., Wang, D., Pan, L., Song, Z., Freitag, M., Wang, W. Y., & Li, L. (2023). Instructscore: Towards explainable text generation evaluation with automatic feedback. arXiv preprint arXiv:2305.14282. <br>\n [14] Majumder, B. P., He, Z., & McAuley, J. (2022). InterFair: Debiasing with Natural Language Feedback for Fair Interpretable Predictions. arXiv preprint arXiv:2210.07440. <br>\n [15] Yang, Z., Feng, R., Zhang, H., Shen, Y., Zhu, K., Huang, L., ... & Cheng, F. (2023). Eliminating Lipschitz Singularities in Diffusion Models. arXiv preprint arXiv:2306.11251. <br>\n [16] Li, J., Zhang, M., Guo, P., Zhang, M., & Zhang, Y. (2023). LLM-enhanced Self-training for Cross-domain Constituency Parsing. arXiv preprint arXiv:2311.02660. <br>\n [17] Ge, S., Zhang, Y., Liu, L., Zhang, M., Han, J., & Gao, J. (2023). Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs. arXiv preprint arXiv:2310.01801. <br>\n [18] Eustratiadis, P., Dudziak, Ł., Li, D., & Hospedales, T. (2023). Neural Fine-Tuning Search for Few-Shot Learning. arXiv preprint arXiv:2306.09295. <br>\n [19] Zhang, Y., Zhang, Y., Cui, L., & Fu, G. (2023). Non-autoregressive text editing with copy-aware latent alignments. arXiv preprint arXiv:2310.07821. <br>\n [20] Tu, H., Li, Y., Mi, F., & Yang, Z. (2023). ReSee: Responding through Seeing Fine-grained Visual Knowledge in Open-domain Dialogue. arXiv preprint arXiv:2305.13602. <br>\n [21] Deng, Y., Zhang, W., Pan, S. J., & Bing, L. (2023). SOUL: Towards Sentiment and Opinion Understanding of Language. arXiv preprint arXiv:2310.17924. <br>\n [22] Singh, G., Ghosh, S., Verma, A., Painkra, C., & Ekbal, A. (2023, December). Standardizing Distress Analysis: Emotion-Driven Distress Identification and Cause Extraction (DICE) in Multimodal Online Posts. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (pp. 4517-4532). <br>\n [23] Cao, Q., Kojima, T., Matsuo, Y., & Iwasawa, Y. (2023, December). Unnatural Error Correction: GPT-4 Can Almost Perfectly Handle Unnatural Scrambled Text. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (pp. 8898-8913). <br>\n [24] Cao, H., Yuan, L., Zhang, Y., & Ng, H. T. (2023, December). Unsupervised Grammatical Error Correction Rivaling Supervised Methods. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (pp. 3072-3088). <br>\n [25] Xu, S., Staufer, L., Ichim, O., Heri, C., & Grabmair, M. (2023). Vechr: A dataset for explainable and robust classification of vulnerability type in the european court of human rights. arXiv preprint arXiv:2310.11368. <br>\n [26] Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., ... & Polosukhin, I. (2017). Attention is all you need. Advances in neural information processing systems, 30. <br>\n [27] Joulin, A., Grave, E., Bojanowski, P., & Mikolov, T. (2016). Bag of tricks for efficient text classification. arXiv preprint arXiv:1607.01759. <br>\n [28] Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. (2018). Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805. <br>\n [29] Sarzynska-Wawer, J., Wawer, A., Pawlak, A., Szymanowska, J., Stefaniak, I., Jarkiewicz, M., & Okruszek, L. (2021). Detecting formal thought disorder by deep contextualized word representations. Psychiatry Research, 304, 114135. <br>\n [30] He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778). <br>\n [31] Sanh, V., Debut, L., Chaumond, J., & Wolf, T. (2019). DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter. arXiv preprint arXiv:1910.01108. <br>\n [32] Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., ... & Liu, P. J. (2020). Exploring the limits of transfer learning with a unified text-to-text transformer. The Journal of Machine Learning Research, 21(1), 5485-5551. <br>\n [33] Ren, S., He, K., Girshick, R., & Sun, J. (2015). Faster r-cnn: Towards real-time object detection with region proposal networks. Advances in neural information processing systems, 28. <br>\n [34] Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., ... & Bengio, Y. (2014). Generative adversarial nets. Advances in neural information processing systems, 27. <br>\n [35] Pennington, J., Socher, R., & Manning, C. D. (2014, October). Glove: Global vectors for word representation. In Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP) (pp. 1532-1543). <br>\n [36] Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., ... & Rabinovich, A. (2015). Going deeper with convolutions. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 1-9). <br>\n [37] Deng, J., Dong, W., Socher, R., Li, L. J., Li, K., & Fei-Fei, L. (2009, June). Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition (pp. 248-255). Ieee. <br>\n [38] Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems, 25. <br>\n [39] Brown, T., Mann, B., Ryder, N., Subbiah, M., Kaplan, J. D., Dhariwal, P., ... & Amodei, D. (2020). Language models are few-shot learners. Advances in neural information processing systems, 33, 1877-1901. <br>\n [40] Radford, A., Wu, J., Child, R., Luan, D., Amodei, D., & Sutskever, I. (2019). Language models are unsupervised multitask learners. OpenAI blog, 1(8), 9. <br>\n [41] Bahdanau, D., Cho, K., & Bengio, Y. (2014). Neural machine translation by jointly learning to align and translate. arXiv preprint arXiv:1409.0473. <br>\n [42] Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., ... & Stoyanov, V. (2019). Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692. <br>\n [43] Reimers, N., & Gurevych, I. (2019). Sentence-bert: Sentence embeddings using siamese bert-networks. arXiv preprint arXiv:1908.10084. <br>\n [44] Sutskever, I., Vinyals, O., & Le, Q. V. (2014). Sequence to sequence learning with neural networks. Advances in neural information processing systems, 27. <br>\n [45] Dai, Z., Yang, Z., Yang, Y., Carbonell, J., Le, Q. V., & Salakhutdinov, R. (2019). Transformer-xl: Attentive language models beyond a fixed-length context. arXiv preprint arXiv:1901.02860. <br>\n [46] Ronneberger, O., Fischer, P., & Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18 (pp. 234-241). Springer International Publishing. <br>\n [47] Howard, J., & Ruder, S. (2018). Universal language model fine-tuning for text classification. arXiv preprint arXiv:1801.06146. <br>\n [48] Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556. <br>\n [49] Yang, Z., Dai, Z., Yang, Y., Carbonell, J., Salakhutdinov, R. R., & Le, Q. V. (2019). Xlnet: Generalized autoregressive pretraining for language understanding. Advances in neural information processing systems, 32. <br>\n [50] Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 779-788). <br>\n</details>" ]

[ "TAGS\n#task_categories-question-answering #language-English #license-mit #arxiv-2402.09910 #region-us \n", "# arXivTection Dataset\nThe arXivTection dataset serves as a benchmark designed for the task of detecting pretraining data from Large Language models.\n\nThe dataset consists of 50 research papers extracted from arXiv. \n- 25 published in 2023: Non-Training data, \"_label_\" column = 0.\n- 25 published before 2022: Training data, \"_label_\" column = 1.\n\nFrom each paper ≈ 30 passages are extracted. Each passage is paraphrased 3 times using the Language Model Claude v2.0. <br>\nThe \"_Answer_\" column indicates which of the passages is the real excerpt.<br>\nPassages are extracted to be on average ≈ 128 tokens in length.\n\n<br>", "# Testing Models on arXivTection\nOur dataset is planned to be used on a Multiple-Choice-Question-Answering format. Nonetheless, it is compatible to be used with other pretraining data detection methods.<br>\nOur GitHub repository contains example scripts to evaluate models on our dataset.\n\n<br>", "# Compatibility\nThe Multiple-Choice-Question-Answering task with our Dataset is designed to be applied to various models, such as:<br>\n- LLaMA-2\n- Mistral\n- Mixtral\n- Chat-GPT (gpt-3.5-turbo-instruct)\n- GPT-3 (text-davinci-003)\n- Claude \n\n<br>", "# Loading the Dataset\n\n\n\n<br>", "# Citation\n\n\n\n<details>\n <summary> Research Papers Used</summary>\n [1] Attanasio, G., Plaza-del-Arco, F. M., Nozza, D., & Lauscher, A. (2023). A Tale of Pronouns: Interpretability Informs Gender Bias Mitigation for Fairer Instruction-Tuned Machine Translation. arXiv preprint arXiv:2310.12127. <br>\n [2] Shi, Y., Wu, L., & Shao, M. (2023). Adaptive End-to-End Metric Learning for Zero-Shot Cross-Domain Slot Filling. arXiv preprint arXiv:2310.15294. <br>\n [3] Keleg, A., Goldwater, S., & Magdy, W. (2023). ALDi: Quantifying the arabic level of dialectness of text. arXiv preprint arXiv:2310.13747. <br>\n [4] Su, Y., Ji, Y., Li, J., Ye, H., & Zhang, M. (2023, December). Beware of Model Collapse! Fast and Stable Test-time Adaptation for Robust Question Answering. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (pp. 12998-13011). <br>\n [5] Chang, Y., Lo, K., Goyal, T., & Iyyer, M. (2023). BooookScore: A systematic exploration of book-length summarization in the era of LLMs. arXiv preprint arXiv:2310.00785. <br>\n [6] Karamolegkou, A., Li, J., Zhou, L., & Søgaard, A. (2023). Copyright Violations and Large Language Models. arXiv preprint arXiv:2310.13771. <br>\n [7] Weissweiler, L., Hofmann, V., Kantharuban, A., Cai, A., Dutt, R., Hengle, A., ... & Mortensen, D. R. (2023). Counting the Bugs in ChatGPT's Wugs: A Multilingual Investigation into the Morphological Capabilities of a Large Language Model. arXiv preprint arXiv:2310.15113. <br>\n [8] Li, Z., & Zhang, Y. (2023, December). Cultural Concept Adaptation on Multimodal Reasoning. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (pp. 262-276). <br>\n [9] Jiang, W., Mao, Q., Li, J., Lin, C., Yang, W., Deng, T., & Wang, Z. (2023). DisCo: Distilled Student Models Co-training for Semi-supervised Text Mining. arXiv preprint arXiv:2305.12074. <br>\n [10] Zhu, Y., Si, J., Zhao, Y., Zhu, H., Zhou, D., & He, Y. (2023). EXPLAIN, EDIT, GENERATE: Rationale-Sensitive Counterfactual Data Augmentation for Multi-hop Fact Verification. arXiv preprint arXiv:2310.14508. <br>\n [11] Hada, R., Seth, A., Diddee, H., & Bali, K. (2023). ''Fifty Shades of Bias'': Normative Ratings of Gender Bias in GPT Generated English Text. arXiv preprint arXiv:2310.17428. <br>\n [12] Song, Y., & Dhariwal, P. (2023). Improved techniques for training consistency models. arXiv preprint arXiv:2310.14189. <br>\n [13] Xu, W., Wang, D., Pan, L., Song, Z., Freitag, M., Wang, W. Y., & Li, L. (2023). Instructscore: Towards explainable text generation evaluation with automatic feedback. arXiv preprint arXiv:2305.14282. <br>\n [14] Majumder, B. P., He, Z., & McAuley, J. (2022). InterFair: Debiasing with Natural Language Feedback for Fair Interpretable Predictions. arXiv preprint arXiv:2210.07440. <br>\n [15] Yang, Z., Feng, R., Zhang, H., Shen, Y., Zhu, K., Huang, L., ... & Cheng, F. (2023). Eliminating Lipschitz Singularities in Diffusion Models. arXiv preprint arXiv:2306.11251. <br>\n [16] Li, J., Zhang, M., Guo, P., Zhang, M., & Zhang, Y. (2023). LLM-enhanced Self-training for Cross-domain Constituency Parsing. arXiv preprint arXiv:2311.02660. <br>\n [17] Ge, S., Zhang, Y., Liu, L., Zhang, M., Han, J., & Gao, J. (2023). Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs. arXiv preprint arXiv:2310.01801. <br>\n [18] Eustratiadis, P., Dudziak, Ł., Li, D., & Hospedales, T. (2023). Neural Fine-Tuning Search for Few-Shot Learning. arXiv preprint arXiv:2306.09295. <br>\n [19] Zhang, Y., Zhang, Y., Cui, L., & Fu, G. (2023). Non-autoregressive text editing with copy-aware latent alignments. arXiv preprint arXiv:2310.07821. <br>\n [20] Tu, H., Li, Y., Mi, F., & Yang, Z. (2023). ReSee: Responding through Seeing Fine-grained Visual Knowledge in Open-domain Dialogue. arXiv preprint arXiv:2305.13602. <br>\n [21] Deng, Y., Zhang, W., Pan, S. J., & Bing, L. (2023). SOUL: Towards Sentiment and Opinion Understanding of Language. arXiv preprint arXiv:2310.17924. <br>\n [22] Singh, G., Ghosh, S., Verma, A., Painkra, C., & Ekbal, A. (2023, December). Standardizing Distress Analysis: Emotion-Driven Distress Identification and Cause Extraction (DICE) in Multimodal Online Posts. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (pp. 4517-4532). <br>\n [23] Cao, Q., Kojima, T., Matsuo, Y., & Iwasawa, Y. (2023, December). Unnatural Error Correction: GPT-4 Can Almost Perfectly Handle Unnatural Scrambled Text. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (pp. 8898-8913). <br>\n [24] Cao, H., Yuan, L., Zhang, Y., & Ng, H. T. (2023, December). Unsupervised Grammatical Error Correction Rivaling Supervised Methods. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (pp. 3072-3088). <br>\n [25] Xu, S., Staufer, L., Ichim, O., Heri, C., & Grabmair, M. (2023). Vechr: A dataset for explainable and robust classification of vulnerability type in the european court of human rights. arXiv preprint arXiv:2310.11368. <br>\n [26] Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., ... & Polosukhin, I. (2017). Attention is all you need. Advances in neural information processing systems, 30. <br>\n [27] Joulin, A., Grave, E., Bojanowski, P., & Mikolov, T. (2016). Bag of tricks for efficient text classification. arXiv preprint arXiv:1607.01759. <br>\n [28] Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. (2018). Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805. <br>\n [29] Sarzynska-Wawer, J., Wawer, A., Pawlak, A., Szymanowska, J., Stefaniak, I., Jarkiewicz, M., & Okruszek, L. (2021). Detecting formal thought disorder by deep contextualized word representations. Psychiatry Research, 304, 114135. <br>\n [30] He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778). <br>\n [31] Sanh, V., Debut, L., Chaumond, J., & Wolf, T. (2019). DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter. arXiv preprint arXiv:1910.01108. <br>\n [32] Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., ... & Liu, P. J. (2020). Exploring the limits of transfer learning with a unified text-to-text transformer. The Journal of Machine Learning Research, 21(1), 5485-5551. <br>\n [33] Ren, S., He, K., Girshick, R., & Sun, J. (2015). Faster r-cnn: Towards real-time object detection with region proposal networks. Advances in neural information processing systems, 28. <br>\n [34] Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., ... & Bengio, Y. (2014). Generative adversarial nets. Advances in neural information processing systems, 27. <br>\n [35] Pennington, J., Socher, R., & Manning, C. D. (2014, October). Glove: Global vectors for word representation. In Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP) (pp. 1532-1543). <br>\n [36] Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., ... & Rabinovich, A. (2015). Going deeper with convolutions. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 1-9). <br>\n [37] Deng, J., Dong, W., Socher, R., Li, L. J., Li, K., & Fei-Fei, L. (2009, June). Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition (pp. 248-255). Ieee. <br>\n [38] Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems, 25. <br>\n [39] Brown, T., Mann, B., Ryder, N., Subbiah, M., Kaplan, J. D., Dhariwal, P., ... & Amodei, D. (2020). Language models are few-shot learners. Advances in neural information processing systems, 33, 1877-1901. <br>\n [40] Radford, A., Wu, J., Child, R., Luan, D., Amodei, D., & Sutskever, I. (2019). Language models are unsupervised multitask learners. OpenAI blog, 1(8), 9. <br>\n [41] Bahdanau, D., Cho, K., & Bengio, Y. (2014). Neural machine translation by jointly learning to align and translate. arXiv preprint arXiv:1409.0473. <br>\n [42] Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., ... & Stoyanov, V. (2019). Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692. <br>\n [43] Reimers, N., & Gurevych, I. (2019). Sentence-bert: Sentence embeddings using siamese bert-networks. arXiv preprint arXiv:1908.10084. <br>\n [44] Sutskever, I., Vinyals, O., & Le, Q. V. (2014). Sequence to sequence learning with neural networks. Advances in neural information processing systems, 27. <br>\n [45] Dai, Z., Yang, Z., Yang, Y., Carbonell, J., Le, Q. V., & Salakhutdinov, R. (2019). Transformer-xl: Attentive language models beyond a fixed-length context. arXiv preprint arXiv:1901.02860. <br>\n [46] Ronneberger, O., Fischer, P., & Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18 (pp. 234-241). Springer International Publishing. <br>\n [47] Howard, J., & Ruder, S. (2018). Universal language model fine-tuning for text classification. arXiv preprint arXiv:1801.06146. <br>\n [48] Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556. <br>\n [49] Yang, Z., Dai, Z., Yang, Y., Carbonell, J., Salakhutdinov, R. R., & Le, Q. V. (2019). Xlnet: Generalized autoregressive pretraining for language understanding. Advances in neural information processing systems, 32. <br>\n [50] Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 779-788). <br>\n</details>" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"full_cleaned_xsum_faith\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"full_cleaned_xsum_faith\"\n\nMore Information needed" ]

[ "task_categories:question-answering", "language:es", "region:us" ]

[]

[ "es" ]

[ "# Dataset Card for Dataset Name\n\n\n\nThis dataset card aims to be a base template for new datasets. It has been generated using this raw template.", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "TAGS\n#task_categories-question-answering #language-Spanish #region-us \n", "# Dataset Card for Dataset Name\n\n\n\nThis dataset card aims to be a base template for new datasets. It has been generated using this raw template.", "## Dataset Details", "### Dataset Description\n\n\n\n\n\n- Curated by: \n- Funded by [optional]: \n- Shared by [optional]: \n- Language(s) (NLP): \n- License:", "### Dataset Sources [optional]\n\n\n\n- Repository: \n- Paper [optional]: \n- Demo [optional]:", "## Uses", "### Direct Use", "### Out-of-Scope Use", "## Dataset Structure", "## Dataset Creation", "### Curation Rationale", "### Source Data", "#### Data Collection and Processing", "#### Who are the source data producers?", "### Annotations [optional]", "#### Annotation process", "#### Who are the annotators?", "#### Personal and Sensitive Information", "## Bias, Risks, and Limitations", "### Recommendations\n\n\n\nUsers should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.\n\n[optional]\n\n\n\nBibTeX:\n\n\n\nAPA:", "## Glossary [optional]", "## More Information [optional]", "## Dataset Card Authors [optional]", "## Dataset Card Contact" ]

[ "task_categories:text-classification", "size_categories:n<1K", "region:us" ]

[]

[]

[]

[ "TAGS\n#task_categories-text-classification #size_categories-n<1K #region-us \n" ]

[ "task_categories:image-to-text", "language:ur", "license:cc-by-nc-4.0", "ocr", "text recognition", "urdu-ocr", "utrnet", "arxiv:2306.15782", "region:us" ]

[ "2306.15782" ]

[ "ur" ]

[]

[ "TAGS\n#task_categories-image-to-text #language-Urdu #license-cc-by-nc-4.0 #ocr #text recognition #urdu-ocr #utrnet #arxiv-2306.15782 #region-us \n" ]

[ "task_categories:image-to-text", "language:ur", "license:cc-by-nc-4.0", "ocr", "text recognition", "urdu-ocr", "utrnet", "arxiv:2306.15782", "region:us" ]

[ "2306.15782" ]

[ "ur" ]

[]

[ "TAGS\n#task_categories-image-to-text #language-Urdu #license-cc-by-nc-4.0 #ocr #text recognition #urdu-ocr #utrnet #arxiv-2306.15782 #region-us \n" ]

[ "task_categories:image-to-text", "language:ur", "license:cc-by-nc-4.0", "ocr", "text recognition", "urdu-ocr", "utrnet", "arxiv:2306.15782", "region:us" ]

[ "2306.15782" ]

[ "ur" ]

[]

[ "TAGS\n#task_categories-image-to-text #language-Urdu #license-cc-by-nc-4.0 #ocr #text recognition #urdu-ocr #utrnet #arxiv-2306.15782 #region-us \n" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"translations\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"translations\"\n\nMore Information needed" ]

[ "region:us" ]

[]

[]

[]

[ "TAGS\n#region-us \n" ]

[ "region:us" ]

[]

[]

[ "# Dataset Card for \"Train_VIVOS_CommonVoice_FOSD_Control_dataset\"\n\nMore Information needed" ]

[ "TAGS\n#region-us \n", "# Dataset Card for \"Train_VIVOS_CommonVoice_FOSD_Control_dataset\"\n\nMore Information needed" ]

[ "size_categories:1K<n<10K", "language:en", "license:mit", "membership inference", "privacy", "arxiv:2402.07841", "region:us" ]

[ "2402.07841" ]

[ "en" ]

[ "# MIMIR\n\nThese datasets serve as a benchmark designed to evaluate membership inference attack (MIA) methods, specifically in detecting pretraining data from extensive large language models.", "## Applicability\n\nThe datasets can be applied to any model trained on The Pile, including (but not limited to):\n- GPTNeo\n- Pythia\n- OPT", "## Loading the datasets\nTo load the dataset:\n\n\n\n- Available Names: 'arxiv', 'dm_mathematics', 'github', 'hackernews', 'pile_cc', 'pubmed_central', 'wikipedia_(en)', 'full_pile', 'c4', 'temporal_arxiv', 'temporal_wiki'\n- Available Splits: 'ngram_7_0.2', 'ngram_13_0.2', 'ngram_13_0.8' (for most sources), 'none' (for other sources)\n- Available Features: 'member' (str), 'nonmember' (str), 'member_neighbors' (List[str]), 'nonmember_neighbors' (List[str])", "## ️ Codebase\nFor evaluating MIA methods on our datasets, visit our GitHub repository.", "## ⭐ Citing our Work\n\nIf you find our codebase and datasets beneficial, kindly cite our work:" ]

[ "TAGS\n#size_categories-1K<n<10K #language-English #license-mit #membership inference #privacy #arxiv-2402.07841 #region-us \n", "# MIMIR\n\nThese datasets serve as a benchmark designed to evaluate membership inference attack (MIA) methods, specifically in detecting pretraining data from extensive large language models.", "## Applicability\n\nThe datasets can be applied to any model trained on The Pile, including (but not limited to):\n- GPTNeo\n- Pythia\n- OPT", "## Loading the datasets\nTo load the dataset:\n\n\n\n- Available Names: 'arxiv', 'dm_mathematics', 'github', 'hackernews', 'pile_cc', 'pubmed_central', 'wikipedia_(en)', 'full_pile', 'c4', 'temporal_arxiv', 'temporal_wiki'\n- Available Splits: 'ngram_7_0.2', 'ngram_13_0.2', 'ngram_13_0.8' (for most sources), 'none' (for other sources)\n- Available Features: 'member' (str), 'nonmember' (str), 'member_neighbors' (List[str]), 'nonmember_neighbors' (List[str])", "## ️ Codebase\nFor evaluating MIA methods on our datasets, visit our GitHub repository.", "## ⭐ Citing our Work\n\nIf you find our codebase and datasets beneficial, kindly cite our work:" ]