让棋类大模型像人一样下棋，而非只追求最强水平。
Tracking vs. Deciding: The Dual-Capability Bottleneck in Searchless Chess Transformers
- 用走法序列训练，需同时掌握棋局追踪和决策质量
- 120M参数模型无搜索下达Lichess闪电赛2570分，人类走法预测准确率55.2%
- 通过加权训练平衡多样性与决策质量，避免过拟合
一个类人棋类引擎应模仿强人类玩家的风格、失误与一致性，而非仅追求最高强度。我们发现，仅从走法序列训练会迫使模型学习两种能力：状态追踪（从走法历史重建棋盘）和决策质量（从重建状态选择好步）。二者数据需求矛盾：低分局提供追踪所需多样性，高分局提供决策质量信号。移除低分局会降低性能。我们将此张力形式化为双能力瓶颈，即整体表现受限于较弱的能力，即 P ≤ min(T, Q)。基于此，我们把模型从28M扩展到120M以增强追踪，并引入埃洛加权训练提升决策同时保留多样性。两因素消融实验表明：扩展提升追踪，加权改善决策，两者组合具有超加性效果。线性加权最优，过度激进加权虽降低验证损失却损害追踪。我们还提出覆盖率衰减公式 t* = log(N/kcrit)/log b，作为对局内退化风险的可靠性时域。最终120M参数模型在无搜索情况下，在253场有等级分对局中达到Lichess闪电赛2570分。在人类走法预测任务上，其Top-1准确率达55.2%，超过Maia-2快速与闪击版本。与基于位置的方法不同，序列输入天然编码完整对局历史，支持依赖历史的决策，这是单位置模型无法实现的。
原文摘要 · Abstract (English)
A human-like chess engine should mimic the style, errors, and consistency of a strong human player rather than maximize playing strength. We show that training from move sequences alone forces a model to learn two capabilities: state tracking, which reconstructs the board from move history, and decision quality, which selects good moves from that reconstructed state. These impose contradictory data requirements: low-rated games provide the diversity needed for tracking, while high-rated games provide the quality signal for decision learning. Removing low-rated data degrades performance. We formalize this tension as a dual-capability bottleneck, P <= min(T,Q), where overall performance is limited by the weaker capability. Guided by this view, we scale the model from 28M to 120M parameters to improve tracking, then introduce Elo-weighted training to improve decisions while preserving diversity. A 2 x 2 factorial ablation shows that scaling improves tracking, weighting improves decisions, and their combination is superadditive. Linear weighting works best, while overly aggressive weighting harms tracking despite lower validation loss. We also introduce a coverage-decay formula, t* = log(N/kcrit)/log b, as a reliability horizon for intra-game degeneration risk. Our final 120M-parameter model, without search, reached Lichess bullet 2570 over 253 rated games. On human move prediction it achieves 55.2% Top-1 accuracy, exceeding Maia-2 rapid and Maia-2 blitz. Unlike position-based methods, sequence input naturally encodes full game history, enabling history-dependent decisions that single-position models cannot exhibit.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。