AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Chessformer: A Unified Architecture for Chess Modeling

Daniel Monroe, George Eilender, Philip Chalmers, Zhenwei Tang, Ashton Anderson

ICLR 2026 · Architectures and efficiency › Attention and transformers · approach: Transformer block and depth design · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace absolute position embedding with GAB at ~4M params (ablation)Chessformer (Leela distillation) · Leela distillation ablation (searchless)Elo: 0 -> 83Improved
replace absolute position embedding with GAB (ablation)Chessformer (Leela distillation) · Leela distillation ablation (searchless)policy accuracy (%): 56.6 -> 58.5Improved
replace absolute position embedding with GAB (ablation)Chessformer (Leela distillation) · Leela distillation ablation (searchless)puzzle accuracy (%): 61.0 -> 64.2Improved
replace absolute position embedding with GAB (ablation)Chessformer (Leela distillation) · Leela distillation ablation (searchless)value accuracy (%): 88.7 -> 89.0Improved
replace relative position bias with GAB (ablation)Chessformer (Leela distillation) · Leela distillation ablation (searchless)Elo: 40 -> 83Improved
use GAB at 2.5M params versus absolute encoding at 3.67M (ablation)Chessformer (Leela distillation) · Leela distillation ablation (searchless)Elo: 0 -> -4No clear effect
replace absolute position embedding with GAB (ablation)Maia-3 (5M) · Human move matching (5M scale)policy accuracy (%): 54.7 -> 55.4Improved
replace relative position bias with GAB (ablation)Maia-3 (5M) · Human move matching (5M scale)policy accuracy (%): 54.6 -> 55.4Improved
use 3M GAB model versus 4.58M absolute-encoding model (ablation)Maia-3 · Human move matching (5M scale)policy accuracy (%): 54.7 -> 54.8No clear effect
add 7 past board positions as input (ablation)Maia-3-5M · Human move matching, history ablationmove-matching accuracy (%): 54.0 -> 55.4Improved
increase history from 7 to 31 past positions (ablation)Maia-3-5M · Human move matching, history ablationmove-matching accuracy (%): 55.4 -> 55.4No clear effect
replace CNN with Chessformer (Leela-CF) in Leela Chess ZeroLeela Chess Zero · Full engine vs Leela-CNNElo: +112 / +111 / +105 Elo at 160k / 320k / 640kImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy