ACT-Eval decomposes chess commentary into atomic claims, verifies them with deterministic chess tools, and shows that tool access roughly halves factual error rates in LLM chess commentary while coverage of expert ideas remains low.
O-O is a strong move
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Hallucinations on the Board: Tool-Augmented Evaluation of LLM Chess Commentary
ACT-Eval decomposes chess commentary into atomic claims, verifies them with deterministic chess tools, and shows that tool access roughly halves factual error rates in LLM chess commentary while coverage of expert ideas remains low.