AIPython★ 1,131⑂ 3 forks
xinxuxin/keystone-bench
Clinical Evidence Benchmark
Benchmarks assistant responses by extracting the single fact underlying each answer and testing if the action follows that evidence.
// KEY FEATURES
- Extracts the single fact that each clinical answer depends on from the QA pair.
- Checks if the assistant’s recommended action moves in line with that extracted fact.
- Uses HealthBench dataset to provide a uniform clinical decision‑evidence benchmark.
#ai-safety#benchmark#clinical-decision-support#dataset#evaluation#health#healthbench#large-language-models#llm-evaluation#medical-ai
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