AI Paper Atlas

Research map / Reinforcement learning / RL from feedback

Adaptive Rollout Allocation for Online Reinforcement Learning with Verifiable Rewards

Hieu Trung Nguyen, Bao Nguyen, Wenao Ma, Yuzhi Zhao, Ruifeng She, Viet Anh Nguyen

ICLR 2026 · Reinforcement learning › RL from feedback · approach: GRPO stability and variants · 13 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 VIP rollout allocationRLOO · AIME24 (Qwen2.5-Math-1.5B, 8xQ)Pass@32: +12.3 (18.29 -> 30.55)Improved
add VIP rollout allocationRLOO · AIME24 (Qwen2.5-Math-1.5B, 8xQ)Mean@32: +6.3 (3.43 -> 9.68)Improved
add VIP rollout allocationDr. GRPO · AIME24 (Qwen2.5-Math-1.5B, 8xQ)Pass@32: 34.56 -> 33.63Hurt
add VIP rollout allocationDr. GRPO · AIME24 (Qwen2.5-Math-7B, 8xQ)Pass@32: 50.0 -> 58.98Improved
add VIP rollout allocationRLOO · AIME24 (Llama-3.2-3B-Instruct, 8xQ)Pass@32: 28.84 -> 35.59Improved
add VIP rollout allocationDr. GRPO · Bamboogle (Qwen2.5-3B-Instruct)EM: 20 -> 23.2Improved
add VIP rollout allocationRLOO · Bamboogle (Qwen2.5-3B-Instruct)EM: 10.4 -> 17.6Improved
add VIP rollout allocationDr. GRPO · MuSiQue test (Qwen2.5-3B-Instruct)EM: 6 -> 10.5Improved
add VIP rollout allocationRLOO · Bamboogle (Qwen2.5-3B-Instruct)F1@5: 0.190 -> 0.264Improved
add adaptive variance-aware allocator instead of Inverse Accuracy heuristic (ablation)RLOO+GP · AIME24 (Qwen2.5-Math-1.5B)Pass@32: 24.61 -> 30.55Improved
add adaptive variance-aware allocator instead of Inverse Variance heuristic (ablation)RLOO+GP · AIME24 (Qwen2.5-Math-1.5B)Mean@32: 4.90 -> 9.68Improved
add Gaussian process predictor instead of Ridge Regression (ablation)RLOO+VIP · AIME24 (Qwen2.5-Math-1.5B)Pass@32: 29.90 -> 30.55Improved

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