checkmate-factbench
MITCheckmate FactBench is a factual accuracy evaluation framework that benchmarks LLMs using the FEVER dataset. It measures how well a model can classify claims (Supported / Refuted / NEI) and cite the correct evidence from Wikipedia.
100
Security score
0 known advisories in v0.2.0
Weekly downloads
—
Unpacked size
986.6 kB
Dependencies
9
Last publish
8 months ago
Security advisories
No known vulnerabilities affect v0.2.0.