AI Paper Atlas

Research map / Reinforcement learning / RL from feedback

Shuffle-R1: Efficient RL framework for Multimodal Large Language Models via Data-centric Dynamic Shuffle

Linghao Zhu, Yiran Guan, Dingkang Liang, Jianzhong Ju, Zhenbo Luo, Bin Qin, Jian Luan, Yuliang Liu, Xiang Bai

ICLR 2026 · Reinforcement learning › RL from feedback · approach: RL for LLM reasoning

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 Pairwise Trajectory Sampling (PTS) to GRPO (ablation)GRPO on Qwen2.5-VL-3B · Geo3K testaccuracy: 42.64 -> 46.21Improved
add Advantage-based Batch Shuffle (ABS) on top of PTS (ablation)Shuffle-R1 on Qwen2.5-VL-3B · Geo3K testaccuracy: 46.21 -> 47.88Improved
apply ABS to uniformly random trajectory pairs without PTS (ablation)GRPO on Qwen2.5-VL-3B · Geo3K testaccuracy: 42.64 -> 46.82Improved
select only highest-advantage trajectories (one-way positive) (ablation)PTS · Geo3K testaccuracy: 42.64 -> 41.26Hurt
select only lowest-advantage trajectories (one-way negative) (ablation)PTS · Geo3K testaccuracy: 42.64 -> 23.36Hurt
use max-min contrastive pairing instead of random trajectory selection (ablation)PTS · Geo3K testaccuracy: 43.53 -> 46.21Improved
use advantage-weighted shuffling instead of uniform shuffle (ablation)ABS · Geo3K testaccuracy: 46.05 -> 47.88Improved
static reorder of batch without sub-sampling (ablation)PTS-only · Geo3K testaccuracy: 46.21 -> 46.28No clear effect
lower PTS sampling ratio alpha from 0.5 to 0.25 (ablation)Shuffle-R1 · Geo3K testaccuracy: 47.88 -> 45.75Hurt
raise PTS sampling ratio alpha from 0.5 to 0.75 (ablation)Shuffle-R1 · Geo3K testaccuracy: 47.88 -> 47.41Hurt
increase ABS shuffle rounds S from 8 to 16 (ablation)Shuffle-R1 · Geo3K testaccuracy: 47.88 -> 47.23Hurt
replace ABS with prioritized experience replay (ablation)Shuffle-R1 · Geo3K training/validation accuracyaccuracyHurt

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