AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Process Reward Agents for Steering Knowledge-Intensive Reasoning

Jiwoong Sohn, Tomasz Sternal, Kenneth Styppa, Torsten Hoefler, Michael Moor

ICML 2026 · Large language models › Retrieval and knowledge · approach: Long-context and personalized retrieval · 0 citations (OpenAlex)

This paper tried a common change: Chain-of-thought prompting

add Self-Consistency to CoT on hard benchmarks · GPQA · accuracy: 60.8 -> 51.8 Hurt

What 1,107 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

Read the paper

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
apply step-wise rewards online during beam search instead of post hoc scoring (ablation)PRA · MedQA testaccuracy: 77.6 -> 81.9Improved
train the reward agent on teacher step labels (ablation)PRA with search · MedQA testaccuracy: 76.7 -> 81.9Improved
add retrieval (search) to an untrained reward agent (ablation)PRA beam search · MedQA testaccuracy: 74.4 -> 76.7Improved
use untrained reward agent in beam search without training (ablation)Qwen3-4B-Instruct policy · MedQA testaccuracy: 74.8 -> 74.4No clear effect
use outcome-level reward (last step) instead of online process rewards (ablation)trained PRA · MedQA testaccuracy: 81.9 -> 75.7Hurt
aggregate process rewards with min post hoc (ablation)trained PRA · MedQA testaccuracy: 74.8 -> 74.3Hurt
add PRA guidance versus Self-Consistency across unseen policy modelsfrozen policies 0.5B to 8B · MedQA testaccuracy: Qwen2.5-0.5B: 28.4 -> 54.1Improved
increase number of sampled traces from 8 to 64 (ablation)Self-Consistency · MedQA testaccuracyNo clear effect
increase number of sampled traces (inference-time scaling) (ablation)PRA · MedQA testaccuracyImproved
add Self-Consistency to CoT on hard benchmarksQwen3-4B-Instruct · GPQAaccuracy: 60.8 -> 51.8Hurt
retrieve selectively using learned search score instead of at every step (ablation)PRA · MedQA testaccuracyMixed

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