LLMs Know More Than They Show: On the Intrinsic Representation of LLM Hallucinations
Paper
•
2410.02707
•
Published
•
49
•
5
Hadas Orgad
hadasor
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
about 2 months ago
Inside-Out: Hidden Factual Knowledge in LLMs
upvoted
a
paper
4 months ago
TokenVerse: Versatile Multi-concept Personalization in Token Modulation
Space
authored
a paper
4 months ago
Padding Tone: A Mechanistic Analysis of Padding Tokens in T2I Models
Organizations
hadasor's activity
commented
2
papers
7 months ago