arXiv cs.CL· Taiyi Chen, Iskar Deng, Amandalynne Paullada·· 12 小时前AI 评分18
用受控评测任务检验语言围棋智能体能否识别 Atari 与评估 Sente
Recognizing Atari and Assessing Sente: Controlled Evaluation Tasks for Language-Based Go Agents
AI 导读
研究者构建了围绕围棋中 atari 与 sente 关系的受控评测集，考察局部棋盘变化下的术语识别与不同战略条件下的落子评估。经术语与解说监督微调后的围棋语言模型 LoGos 难以识别 atari 关系的变化并判断其战略含义，其术语预测还与落子偏好相关联，即使这些偏好与参考评估相矛盾。该测试框架可用于检验语言围棋智能体是否将领域术语与精确的棋盘关系和落子价值联系起来。
正文 · 原文
Abstract:Domain terminology provides concise descriptions of game states and strategic relationships, but whether game-playing language models can accurately recognize these concepts and relate them to move evaluation remains unclear. We construct a controlled evaluation set centered on atari and its relationship to sente in Go, examining terminology recognition under local board changes and move evaluation under different strategic conditions. We conduct our tests on LoGos, a Go language model, after fine-tuning it with terminology and commentary supervision. The model struggles to recognize changes in the atari relationship and assess its strategic implications. Its terminology predictions are associated with move preferences even when those preferences contradict the reference evaluation. Our controlled tests provide a framework for examining whether language-based Go agents connect domain terminology with precise board relationships and strategic move values.
| Comments: | 7 pages, 2 figures, 5 tables. The first two authors contributed equally |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.22953 [cs.CL] |
| (or arXiv:2610.22953v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.22953 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Iskar Deng [view email]
[v1]
Fri, 16 Oct 2026 05:23:01 UTC (2,956 KB)
来源：arXiv cs.CL · arxiv.org