From 489932e0aab1e38abfa1258d9ab517b26bb7a451 Mon Sep 17 00:00:00 2001 From: connerlambden Date: Thu, 4 Jun 2026 22:39:32 -0600 Subject: [PATCH] Add BGPT REFUTE benchmark (scientific critique & calibration) --- README.md | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/README.md b/README.md index 2616d5d..d8b402d 100644 --- a/README.md +++ b/README.md @@ -131,3 +131,8 @@ Have questions, collaboration ideas, or want to discuss agent evaluation? Reach > Let every agent capability evaluation be evidence-based. Maintained by **DataAnswer (XuChao)** since 2025. + + +## Benchmarks + +- [REFUTE](https://huggingface.co/datasets/BGPT-OFFICIAL/refute) — Scientific critique & epistemic calibration on recent science summaries (Apache-2.0). [Leaderboard](https://huggingface.co/spaces/BGPT-OFFICIAL/refute-leaderboard) · [Technical report](https://huggingface.co/datasets/BGPT-OFFICIAL/refute/blob/main/TECHNICAL_REPORT.md) · [Integrators](https://huggingface.co/datasets/BGPT-OFFICIAL/refute/blob/main/INTEGRATORS.md)