Are LLMs Better than Reported? Detecting Label Errors and Mitigating Their Effect on Model Performance
Paper
• 2410.18889 • Published
• 15
Omer Nahum
omer6nahum
AI & ML interests
None yet
Recent Activity
upvoted a paper 3 days ago
Thinking to Recall: How Reasoning Unlocks Parametric Knowledge in LLMs upvoted a paper 23 days ago
Empty Shelves or Lost Keys? Recall Is the Bottleneck for Parametric Factuality upvoted a paper about 2 months ago
LIBERTy: A Causal Framework for Benchmarking Concept-Based Explanations of LLMs with Structural Counterfactuals Organizations
None yet
authored a paper over 1 year ago
authored a paper almost 2 years ago