跳到正文
arXiv cs.AI· Zhuangsheng Li, Yunhan Zhang, Dailin Li, Qingxian Yu, Yu Wu, Youqi Song, Bangyuan He·· 4 小时前AI 评分29

TReC:评估围棋指令模型中落子引发的状态变化

TReC: Evaluating Move-Induced State Changes in Instruction-Conditioned Go

AI 导读

该论文提出 TReC(Transition Representation Consistency),用于比较围棋模型预测落子与专家落子对棋盘状态嵌入的改变:两步棋在同一局面下按规则执行后,TReC 取两次表示变化的余弦相似度。

正文 · 原文

View PDF HTML (experimental)

Abstract:When a Go model predicts a move different from the expert's, coordinate distance measures how far the prediction lands from the reference. We introduce Transition Representation Consistency (TReC) to compare how the two moves change a board-state embedding. Both moves are executed from the same position under the game rules, and TReC is the cosine similarity between the two resulting representation changes. In experiments with a frozen KataGo network on 680 mid-to-late-game positions from five games, the gap between the highest and lowest candidate per position has a median of 0.01 to 0.08 for the direct successor-state cosine and 0.88 to 1.33 for TReC, depending on the evaluator layer. TReC follows coordinate distance only weakly: the rank correlation between a candidate's distance to the reference and its TReC, computed within each position, has a median of -0.19 over these positions.
Subjects: Artificial Intelligence (cs.AI)
Cite as: arXiv:2611.17781 [cs.AI]
  (or arXiv:2611.17781v1 [cs.AI] for this version)
  https://doi.org/10.48550/arXiv.2611.17781

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Zhuangsheng Li [view email]
[v1] Fri, 13 Nov 2026 05:48:21 UTC (159 KB)

来源:arXiv cs.AI · arxiv.org