Published October 16, 2026 | Version v0.2.0

benchmark-honesty

Authors/Creators

Description

Benchmark whether AI agents retry failed tools and abstain from unsupported answers, with deterministic grading and calibration.

Notes

If you use this work, please cite it using the metadata below.

Files

pooriaarab/benchmark-honesty-v0.2.0.zip

Files (63.8 kB)

Name Size
md5:c76eeb73b6fc43ba4df1963988ee3a10
63.8 kB Preview Download

Additional details

Related works