跳到正文
原文
Hugging Face Blog·· 2024-01-29AI 评分36

Hugging Face 发布 Hallucinations Leaderboard,开放评测大语言模型幻觉

The Hallucinations Leaderboard, an Open Effort to Measure Hallucinations in Large Language Models

AI 导读

Hugging Face 推出 Hallucinations Leaderboard,基于 EleutherAI Language Model Evaluation Harness 在 NQ Open、TriviaQA、TruthfulQA、XSum、CNN/DM、RACE、SQuADv2、FEVER、FaithDial、HaluEval 等基准上评测 LLM 的幻觉表现。

来源:Hugging Face Blog · huggingface.co