AIPython★ 3,327⑂ 108 forksApache-2.0
vectara/hallucination-leaderboard
RANKING LLM HALLUCINATIONS
The project compares the performance of large language models in generating hallucinations in short document summaries. It provides a leaderboard with accuracy metrics and the frequency of fabricated facts.
// KEY FEATURES
- Compares LLMs by hallucination rate in document summarization up to 500 tokens.
- Computes metrics: fabricated‑fact %, KL‑divergence, BLEU penalty.
- Leaderboard table with filtering by model, dataset, and hallucination level.
- Python code with CLI, Docker image, and scripts to run evaluation.
New repositories every 30 minutes
REDDYX AI scans GitHub 24/7 and ships the best AI/ML/Web3 projects to Telegram.
Join on Telegram