AI Paper Atlas

Research map / Large language models / Instruction and fine-tuning

How Reasoning Evolves from Post-Training Data: An Empirical Study Using Chess

Lucas Dionisopoulos, Nicklas Majamaki, Prithviraj Ammanabrolu

ICML 2026 · Large language models › Instruction and fine-tuning · approach: Fine-tuning for reasoning data · 0 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
add RL after SFT on Best Move dataSFT6 [BM] · Predict Move% legal moves: 44% -> 93%Improved
add RL after SFT on Best Move - All dataSFT8 [BM - All] · Predict Move% trivial moves: 3% -> 14%Improved
add RL to Best Line SFT checkpointSFT7 [BL] · Predict Move% trivial moves: 2% -> 64%Hurt
add all data types to Best Move SFT mix (ablation)SFT6 [BM] -> SFT8 [BM - All] · Predict Move% trivial moves: 31% -> 14%Improved
add all data types to Best Move SFT mix (ablation)SFT6 [BM] -> SFT8 [BM - All] · Best Move taskaccuracy after RL: 37% -> 41%Improved
use multitask RL (all four tasks) instead of Predict Move only (ablation)SFT2 [RSA] vs SFT1 [RSPM] · Predict Moveaverage normalized rank after RL: 0.52 -> 0.60Improved
add Verbalized Alpha-Beta Pruning SFT data instead of Best Move data (ablation)SFT3 [VABP] vs SFT6 [BM] · Predict Move% legal moves after RL: 93% -> 67%Hurt
scale SFT tokens from 15M to 60M and RL samples from 8k to 16k (ablation)SFT8 [BM - All] -> SFT8 XL · Predict Moveaverage normalized rank after RL: 0.71 -> 0.82Improved
chain SFT9 XL dataset after SFT8 XL checkpoint (ablation)SFT8 XL · Legal Moves taskIoU after RL: 0.79 -> 0.87Improved
add RL to SFT checkpointSFT8 XL · Referenced move accuracy% legal referenced moves: 48% -> 83%Improved
RL on Best Move checkpoint rather than Best Line checkpoint (ablation)SFT6 [BM] · reasoning faithfulnessfaithfulnessHurt
add RL to Qwen2.5-7B average reasoning qualitySFT8 XL · Average reasoning quality (gpt-oss-120b judge)score: 5.0 -> 2.2Hurt

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