AIPython★ 636⑂ 69 forksMIT
HowieHwong/TrustLLM
TRUSTLLM: EVALUATION OF LLM TRUSTWORTHINESS
TrustLLM — a toolkit for assessing LLM trustworthiness, presented at ICML 2024. It includes benchmark suites, metrics, and datasets for measuring hallucinations, bias, and answer reliability.
// KEY FEATURES
- Provides benchmark suites for measuring hallucinations and factuality of LLMs.
- Includes metrics for assessing bias, toxicity, and fairness of model outputs.
- Enables automated evaluation via CLI and API, producing detailed JSON reports.
- Integrates with HuggingFace, LangChain, and other frameworks for quick plug‑and‑play.
#ai#benchmark#dataset#evaluation#large-language-models#llm#natural-language-processing#nlp#pypi-package#toolkit#trustworthy-ai#trustworthy-machine-learning
Open on GitHub →New repositories every 30 minutes
REDDYX AI scans GitHub 24/7 and ships the best AI/ML/Web3 projects to Telegram.
Join on Telegram