NetArtifactBench
Explore score history ↗NetArtifactBench evaluates the ability of AI agents to repair inconsistent network experiment records while preserving claims that remain supported. It contains 52 instances with injected inconsistencies ranging from direct contradictions to unstated relations across multiple artifacts.
- Released
- 2026-09-09
- Institutions
- University of Hong Kong · Tsinghua University · University of California, Santa Cruz
- Evaluates
- General AI, Language & Knowledge, cs.NI
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- ai-reviewed
NetArtifactBench paper
No reported scores are on record for this benchmark yet.