Python
xinxuxin/keystone-bench
A clinical decision-evidence benchmark: pull the one fact an answer rests on, and see whether the assistant's action moves with the evidence. Built on HealthBench.
★ 1.1k
Stars
3 forks · 0 open issues
+18/day
Growth · measured
Stars counted between two of our own snapshots roughly a week apart. The repository page gives the exact span.
47.0
Momentum / 100
see the breakdown below
2mo ago
Created
last push 19d ago
Star history
How this score was built
| Component | Weight | Earned | Why |
|---|---|---|---|
| Velocity | 55 | 24.5 | +18/day, log-scaled against a 750/day ceiling |
| Acceleration | 15 | 0.0 | last week ran 0.98× the monthly pace |
| Freshness | 15 | 13.6 | last push 19d ago; decays to zero at 180 days |
| Scale | 15 | 8.9 | 1.1k stars, log-scaled against 150k — stops a 3-star repo topping the board |
| Momentum | 100 | 47.0 | sum of the above |
Detail
| Topics | ai-safetybenchmarkclinical-decision-supportdatasetevaluationhealthhealthbenchlarge-language-modelsllm-evaluationmedical-ai |
| Matched on | LLM & Inference (large-language-models) |
| Short window | +127 stars over 7 days = 18.1/day |
| Long window | +480 stars over 26 days = 18.5/day |
| Lifetime average | 19.26 stars/day over 59 days |
| First seen here | 2026-10-11 |
| Last read | 2d ago |