Hallucinations on the Board: Tool-Augmented Evaluation of LLM Chess Commentary
S. Ashwin Hebbar, Peiyao Sheng, Sewoong Oh, Pramod Viswanath
cs.CL, cs.AI
Submitted: 2026-08-04
Comments: 23 pages, 6 figures
Code: https://github.com/hebbarashwin/act_eval
License: http://creativecommons.org/licenses/by/4.0/
The gist: Superhuman game engines in domains like chess have made expert-level evaluations easily accessible, yet they communicate what is true without the natural-language explanations that make such
Terminology
Abstract
Superhuman game engines in domains like chess have made expert-level evaluations easily accessible, yet they communicate what is true without the natural-language explanations that make such expertise educationally useful to experts and non-experts alike. Large language models could, in principle, bridge this gap, but they frequently hallucinate due to limited domain-specific knowledge, and standard reference-based or LLM-as-a-judge frameworks cannot reliably detect these errors. In this work, we present ACT-Eval, an evaluation framework that decomposes chess commentary into atomic claims and routes them to engine-supported tools and expert-annotated gold references to assess factual correctness, conceptual coverage, and move-quality judgment. We release a benchmark of 325 position--move pairs spanning pedagogical, tournament, and critical positions, including 125 positions with expert-verified gold atoms and a five-class error taxonomy. Evaluating leading proprietary and open-weight models, we find that factual hallucinations remain pervasive in chess commentary: GPT-5.4 without tools produces incorrect sub-claims 22.0% of the time, while smaller open-weight models exceed 40%. Although tool augmentation substantially improves factual correctness and move-quality assessment, coverage of expert strategic and tactical ideas remains limited across all models. Human calibration shows that ACT-Eval's factual judgments fall within the observed range of inter-human agreement, while its coverage scores correlate strongly with human assessments of strategic completeness.
Sources
- FacTool: Factuality Detection in Generative AI -- A Tool Augmented Framework for Multi-Task and Multi-Domain Scenarios
- Improving Chess Commentaries by Combining Language Models with Symbolic Reasoning Engines
- LLM CHESS: Benchmarking Reasoning and Instruction-Following in LLMs through Chess
- ChessQA: Evaluating Large Language Models for Chess Understanding
- BERTScore: Evaluating Text Generation with BERT
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering