mt5-small-finetune-finetuned-research-papers-XX

This model is a fine-tuned version of google/mt5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.5998
• Rouge1: 36.5831
• Rouge2: 17.8222
• Rougel: 32.0591
• Rougelsum: 32.1426
• Gen Len: 16.0415

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5.6e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
6.6546	0.499	499	3.0145	27.7249	12.6467	24.9735	24.9728	12.4275
3.7935	0.998	998	2.7804	35.8013	17.0498	31.3295	31.3783	15.984
3.4614	1.4970	1497	2.6818	35.7213	17.1243	31.4956	31.552	15.3485
3.2954	1.996	1996	2.6486	35.5961	17.2535	31.3263	31.411	15.857
3.1932	2.495	2495	2.6300	36.5296	17.8491	32.1923	32.2628	15.8925
3.1508	2.9940	2994	2.6121	36.4565	17.6813	32.013	32.0796	16.0425
3.0782	3.493	3493	2.6094	36.4064	17.7208	31.9757	32.0421	16.0315
3.1005	3.992	3992	2.5998	36.5831	17.8222	32.0591	32.1426	16.0415

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.1
• Tokenizers 0.21.1