Hugging Face Trending Papers

Does Bielik Know What It Doesn't Know? Activation Dispersion Separates Entity Familiarity from Factual Reliability Across Model Scale

Read the original on Hugging Face Trending Papers →

Large language models hallucinate most about entities they have never seen. We ask whether a model's activations betray entity familiarity before a single answer token is generated, and whether that signal predicts the factual reliability of the answers.

Summary generated by The Flow from the publisher's feed. The full article lives at Hugging Face Trending Papers.