AI Paper Atlas

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

Beyond Two-Stage Training: Cooperative SFT and RL for LLM Reasoning

Liang Chen, Xueting Han, Li Shen, Jing Bai, Kam-Fai Wong

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 the meta-objective J_meta (bilevel upper-level update) (ablation)BRIDGE on Qwen3-8B-Base · Math average (5 benchmarks)average accuracy: 40.3 -> 49.9Improved
learn SFT guidance via bilevel LoRA teacher instead of naive SFT+RL mixingSFT+RL on Qwen2.5-3B · Math average (5 benchmarks)average accuracy: 29.4 -> 36.0Improved
add SFT warm-up before RL (SFT->RL vs RL)RL on Llama-3.2-3B-Instruct · Math average (5 benchmarks)average accuracy: 21.8 -> 18.9Hurt
add SFT warm-up before RL (SFT->RL vs RL)RL on Qwen3-8B-Base · Math average (5 benchmarks)average accuracy: 42.9 -> 45.5Improved
add SFT warm-up before RL (SFT->RL vs RL)RL on Qwen2.5-3B · Math average (5 benchmarks)average accuracy: 32.2 -> 32.2No clear effect
add periodic SFT updates during RL stage (freq=10) (ablation)SFT->RL on Qwen2.5-3B · Math average (5 benchmarks)average accuracy: 32.2 -> 31.4No clear effect
add periodic SFT updates during RL stage (freq=5) (ablation)SFT->RL on Qwen2.5-3B · Math average (5 benchmarks)average accuracy: 32.2 -> 31.5No clear effect
change mixing weight lambda from 0.5 to 0.0 (pure SFT) (ablation)BRIDGE on Qwen3-8B-Base · Math average (5 benchmarks)average accuracy: 49.9 -> 37.6Hurt
change mixing weight lambda from 0.5 to 1.0 (pure RL) (ablation)BRIDGE on Qwen3-8B-Base · Math average (5 benchmarks)average accuracy: 49.9 -> 42.9Hurt
vary lambda within 0.3-0.7 (ablation)BRIDGE on Qwen3-8B-Base · Math average (5 benchmarks)average accuracy: 47.2--49.9No clear effect
change LoRA rank/alpha from 32/16 to 16/32 (ablation)BRIDGE on Qwen3-8B-Base · Math average (5 benchmarks)average accuracy: 49.9 -> 49.5No clear effect
use BRIDGE instead of RL on Knights-and-KnavesRL on Qwen2.5-3B · Knights-and-Knaves (avg)accuracy: 37.0 -> 55.3Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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