AI Paper Atlas

Research map / Large language models / Agents, code and math

OR-PRM: A Process Reward Model for Algorithmic Problem in Operations Research

Yilin Wang, Heng Zhou, Dongxing Mao, Linjie Li, Jingru Tan, Haochen Han, Zhengyuan Yang, Alex Jinpeng Wang, Min Li

ICLR 2026 · Large language models › Agents, code and math · approach: LLMs for optimization and heuristics

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 OR-PRM Best-of-8 selectionQwen2.5-7B · Overall (6 benchmarks avg)accuracy (%): 34.9 -> 42.9Improved
add OR-PRM Best-of-8 selectionQwen2.5-14B · Overall (6 benchmarks avg)accuracy (%): 48.8 -> 61.3Improved
add OR-PRM Best-of-8 selectionQwen2.5-32B · Overall (6 benchmarks avg)accuracy (%): 59.6 -> 70.3Improved
add OR-PRM Best-of-8 selectionLLMOPT · Overall (6 benchmarks avg)accuracy (%): 73.6 -> 80.5Improved
add OR-PRM Best-of-8 selectionQwen2.5-32B · Complex-LPaccuracy (%): 8.2 -> 32.4Improved
add OR-PRM Best-of-8 selectionQwen2.5-14B · Easy-LPaccuracy (%): 66.2 -> 89.4Improved
add OR-PRM Best-of-8 selectionLLMOPT · Complex-LPaccuracy (%): 48.6 -> 67.6Improved
add OR-PRM Best-of-8 selectionLLMOPT · Easy-LPaccuracy (%): 96.0 -> 97.8Improved
add DPO alignment after SFT (ablation)OR-PRM (Qwen2.5-14B) · Easy-LP and Complex-LP averageaverage accuracy (%): 43.0 -> 51.0Improved
add DPO alignment after SFT (ablation)OR-PRM (Qwen2.5-14B) · Easy-LPaccuracy (%): 79.6 -> 89.4Improved
replace self-consistency majority voting with OR-PRM selectionQwen2.5-14B · Easy-LP and Complex-LP averageaverage accuracy (%): 27.2 -> 51.0Improved
add OR-PRM critic in modeling-critique-code pipelineQwen2.5-14B · Complex-LP pass@1accuracy (%): 3.6 -> 27.0Improved

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