AI Paper Atlas

Research map / Reinforcement learning / RL from feedback

Smarter Not Harder: Generative Process Evaluation with Intrinsic-Signal Driving and Ability‑Adaptive Reward Shaping

Tao He, Rongchuan Mu, Lizi Liao, Yixin Cao, Yang Li, Yijia Luo, Weixun Wang, Ming Liu, Bing Qin

ICLR 2026 · Reinforcement learning › RL from feedback · approach: Scaling RL with verifiable rewards

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 intrinsic-signal-driven evaluation (vs LLM-as-a-judge) (ablation)TP-GRPO · AIME 24 (1.5B)Avg@16 accuracy: 31.04 -> 33.12Improved
add intrinsic-signal-driven evaluation (vs LLM-as-a-judge) (ablation)TP-GRPO · AMC 23 (1.5B)Avg@16 accuracy: 63.93 -> 64.01No clear effect
add thought-level step merging (ablation)TP-GRPO · AIME 25 (1.5B)Avg@16 accuracy: 22.29 -> 25.63Improved
add thought-level step merging (ablation)TP-GRPO · AIME 24 (1.5B)Avg@16 accuracy: 31.66 -> 33.12Improved
add difficulty-aware adaptive reward (vs fixed +1/-1) (ablation)TP-GRPO · AIME 25 (1.5B)Avg@16 accuracy: 22.92 -> 25.63Improved
add difficulty-aware adaptive reward (vs fixed +1/-1) (ablation)TP-GRPO · AIME 24 (1.5B)Avg@16 accuracy: 32.71 -> 33.12Improved
add process rewards for correct solutions (ablation)TP-GRPO · AIME 24 (1.5B)Avg@16 accuracy: 30.00 -> 33.12Improved
add process rewards for incorrect solutions (ablation)TP-GRPO · AIME 25 (1.5B)Avg@16 accuracy: 22.29 -> 25.63Improved
add process rewards for incorrect solutions (ablation)TP-GRPO · AIME 24 (1.5B)Avg@16 accuracy: 32.01 -> 33.12Improved
add process rewards for incorrect solutions (ablation)TP-GRPO · AIME 25 vs AIME 24Avg@16 accuracy: 30.00 -> 33.12Mixed
raise alpha from 0 to 1 (ablation)TP-GRPO · AIME24/AIME25/AMC23 avg (1.5B)average accuracy: 39.17 -> 40.92Improved
raise alpha from 1 to 4 (ablation)TP-GRPO · AIME24/AIME25/AMC23 avg (1.5B)average accuracy: 40.92 -> 39.72Hurt

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