Chessformer: A Unified Architecture for Chess Modeling

• URL: http://arxiv.org/abs/2605.19091v1
• Date: Mon, 18 May 2026 20:27:01 GMT
• Title: Chessformer: A Unified Architecture for Chess Modeling
• Abstract summary: Chess has long served as a canonical testbed for artificial intelligence, but modeling approaches for its central tasks have diverged.<n>We introduce Chessformer, a unified architecture that advances the state of the art on all three central goals in chess modeling.
• Score: 8.811775497383227
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Chess has long served as a canonical testbed for artificial intelligence, but modeling approaches for its central tasks have diverged. Maximizing playing strength, predicting human play, and enabling interpretability are typically solved with disparate architectures, and these designs are often misaligned with the geometry of the domain. This raises the natural question of whether these objectives require separate modeling paradigms, or if there exists a single architecture that supports them simultaneously. We introduce Chessformer, a unified architecture that advances the state of the art on all three central goals in chess modeling. Chessformer is an encoder-only transformer that represents board squares as tokens, augments self-attention with a novel dynamic positional encoding called Geometric Attention Bias (GAB) that adapts to domain-specific geometry, and predicts actions with an attention-based source-destination policy head. We evaluate Chessformer on each front. First, we develop \maiathree, a family of models for human move prediction that reaches 57.1\% move-matching accuracy, significantly surpassing the previous state of the art with fewer than a quarter of the parameters. Second, we integrate Chessformer into Leela Chess Zero, a leading open-source engine, adding over 100 Elo of playing strength and resulting in tournament victories over Stockfish in major computer chess competitions. Third, we show that Chessformer's square-token design makes attention patterns and activations directly attributable to board squares, enabling granular interpretability analyses that prior architectures do not naturally support. More broadly, our results demonstrate that aligning a model's tokenization, positional encoding, and output design with the underlying structure of a domain can yield simultaneous gains in performance, human compatibility, and interpretability.

Related papers

• Parametric Open Source Games [78.29352305480285]
  We introduce parametric open-source games, a continuous analogue of program equilibria.<n>We derive an exact coupling threshold at which selfish gradient ascent in symmetric $2times2$ games switches from defection toward cooperation.<n>We extend the framework to a neural semantics class whose first-order cooperation condition is governed by the ratio of cross-player to self-player sensitivity.
  arXiv  Detail & Related papers  (2026-06-25T14:14:24Z)
• Geometric Action Model for Robot Policy Learning [68.6657929619782]
  Generalist robot policies must follow user instructions while reasoning about how objects, cameras, and robot actions interact in the 3D physical world.<n>Recent vision-language-action models (VLAs) and video world-action models (WAMs) inherit strong semantic or temporal priors from large-scale foundation models.<n>We propose the Geometric Action Model (GAM), a language-conditioned manipulation policy.
  arXiv  Detail & Related papers  (2026-06-15T17:58:03Z)
• RePAIR: Predictive Self-Supervised Representation Learning in Chess [2.1665689529884697]
  We introduce Representation Prediction via Autoencoding using Iterative Refinement (RePAIR)<n>We demonstrate how it can be used to encode objects in sequential data like consecutive chess positions into compact yet meaningful representations.
  arXiv  Detail & Related papers  (2026-06-10T09:36:31Z)
• GenTac: Generative Modeling and Forecasting of Soccer Tactics [66.35532694378146]
  GenTac conceptualizes soccer tactics as a process over continuous multi-player trajectories and discrete semantic events.<n>It supports rich contextual conditioning, including opponent behavior, specific team or league playing styles, and strategic objectives.<n>It can be successfully trained to generalize to other dynamic team sports, including basketball, American football, and ice hockey.
  arXiv  Detail & Related papers  (2026-04-13T17:53:49Z)
• Learning 3D Representations for Spatial Intelligence from Unposed Multi-View Images [81.94999489820974]
  UniSplat is a feed-forward framework for learning 3D representations from unposed multi-view images.<n>We introduce a dual-masking strategy that strengthens geometry induction in the encoder.<n>Second, we develop a coarse-to-fine Gaussian splatting strategy that reduces appearance-semantics inconsistencies.<n>Third, we introduce a pose-conditioned recalibration mechanism that interrelates predicted 3D point and semantic maps into the image plane.
  arXiv  Detail & Related papers  (2026-04-12T10:36:18Z)
• Predicting Human Chess Moves: An AI Assisted Analysis of Chess Games Using Skill-group Specific n-gram Language Models [0.0]
  The framework employs n-gram language models to capture move patterns characteristic of specific player skill levels.<n>We trained separate models using data from the open-source chess platform Lichess.<n>The framework can classify skill levels with an accuracy of up to 31.7% when utilizing early game information.
  arXiv  Detail & Related papers  (2025-12-01T17:02:07Z)
• From Spatial to Actions: Grounding Vision-Language-Action Model in Spatial Foundation Priors [54.84863164684646]
  Existing vision-language-action (VLA) models act in 3D real-world but are typically built on 2D encoders.<n>In this work, we introduce FALCON, a novel paradigm that injects rich 3D spatial tokens into the action head.
  arXiv  Detail & Related papers  (2025-10-20T11:26:45Z)
• Explore the Reasoning Capability of LLMs in the Chess Testbed [45.12891789312405]
  We propose improving the reasoning capability of large language models in chess by integrating annotated strategy and tactic.<n>We finetune the LLaMA-3-8B model and compare it against state-of-the-art commercial language models in the task of selecting better chess moves.
  arXiv  Detail & Related papers  (2024-11-11T01:42:56Z)
• Enhancing Chess Reinforcement Learning with Graph Representation [21.919003715442074]
  We introduce a more general architecture based on Graph Neural Networks (GNN) We show that this new architecture outperforms previous architectures with a similar number of parameters. We also show that the model, when trained on a smaller $5times 5$ variant of chess, is able to be quickly fine-tuned to play on regular $8times 8$ chess.
  arXiv  Detail & Related papers  (2024-10-31T09:18:47Z)
• Mastering Chess with a Transformer Model [0.0]
  We show that transformers endowed with a sufficiently expressive position representation can match existing chess-playing models at a fraction of the computational cost. Our architecture, which we call the Chessformer, significantly outperforms AlphaZero in both playing strength and puzzle solving ability with 8x less computation.
  arXiv  Detail & Related papers  (2024-09-18T19:05:21Z)
• Amortized Planning with Large-Scale Transformers: A Case Study on Chess [11.227110138932442]
  This paper uses chess, a landmark planning problem in AI, to assess performance on a planning task. ChessBench is a large-scale benchmark of 10 million chess games with legal move and value annotations (15 billion points) provided by Stockfish. We show that, although a remarkably good approximation can be distilled into large-scale transformers via supervised learning, perfect distillation is still beyond reach.
  arXiv  Detail & Related papers  (2024-02-07T00:36:24Z)
• Determining Chess Game State From an Image [19.06796946564999]
  This paper puts forth a new dataset synthesised from a 3D model that is an order of magnitude larger than existing ones. A novel end-to-end chess recognition system is presented that combines traditional computer vision techniques with deep learning. The described system achieves an error rate of 0.23% per square on the test set, 28 times better than the current state of the art.
  arXiv  Detail & Related papers  (2021-04-30T13:02:13Z)
• Learning Chess Blindfolded: Evaluating Language Models on State Tracking [69.3794549747725]
  We consider the task of language modeling for the game of chess. Unlike natural language, chess notations describe a simple, constrained, and deterministic domain. We find that transformer language models can learn to track pieces and predict legal moves with high accuracy when trained solely on move sequences.
  arXiv  Detail & Related papers  (2021-02-26T01:16:23Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.