# StepORLM: A Self-Evolving Framework With Generative Process Supervision For Operations Research Language Models

arXiv: https://arxiv.org/abs/2509.22558 · HTML: https://mlsanity.com/paper/2509.22558/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- Placed in Process reward models for reasoning, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 22 of its 29 references that are in the map.
- First posted on arXiv 26 Sep 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Process reward models for reasoning](https://mlsanity.com/alignment/process-reward-models-for-reasoning/)
- Synthetic data and self-improvement for LLMs / [Self-rewarding alignment and self-training](https://mlsanity.com/synth/self-rewarding-alignment-and-self-training/)

## Builds on (most cited first, 22 of 22)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [ReEvo: Large Language Models as Hyper-Heuristics with Reflective Evolution](https://mlsanity.com/paper/2402.01145/) (2024)
- [Evolution of Heuristics: Towards Efficient Automatic Algorithm Design Using Large Language Model](https://mlsanity.com/paper/2401.02051/) (2024)
- [A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems](https://mlsanity.com/paper/2508.07407/) (2025)
- [NL4Opt Competition: Formulating Optimization Problems Based on Their Natural Language Descriptions](https://mlsanity.com/paper/2303.08233/) (2023)
- [ORLM: A Customizable Framework in Training Large Models for Automated Optimization Modeling](https://mlsanity.com/paper/2405.17743/) (2024)
- [OptiMUS: Scalable Optimization Modeling with (MI)LP Solvers and Large Language Models](https://mlsanity.com/paper/2402.10172/) (2024)
- [Monte Carlo Tree Search for Comprehensive Exploration in LLM-Based Automatic Heuristic Design](https://mlsanity.com/paper/2501.08603/) (2025)
- [LLMOPT: Learning to Define and Solve General Optimization Problems from Scratch](https://mlsanity.com/paper/2410.13213/) (2024)
- [A Survey of AI Agent Protocols](https://mlsanity.com/paper/2504.16736/) (2025)
- [A Survey of LLM-based Deep Search Agents: Paradigm, Optimization, Evaluation, and Challenges](https://mlsanity.com/paper/2508.05668/) (2025)
- [Augmenting Operations Research with Auto-Formulation of Optimization Models from Problem Descriptions](https://mlsanity.com/paper/2209.15565/) (2022)
- [Autoformulation of Mathematical Optimization Models Using LLMs](https://mlsanity.com/paper/2411.01679/) (2024)
- [Solver-Informed RL: Grounding Large Language Models for Authentic Optimization Modeling](https://mlsanity.com/paper/2505.11792/) (2025)
- [OptiMUS-0.3: Using Large Language Models to Model and Solve Optimization Problems at Scale](https://mlsanity.com/paper/2407.19633/) (2024)
- [A Survey of Optimization Modeling Meets LLMs: Progress and Future Directions](https://mlsanity.com/paper/2508.10047/) (2025)
- [Retrieval-Augmented Process Reward Model for Generalizable Mathematical Reasoning](https://mlsanity.com/paper/2502.14361/) (2025)
- [Agentic Information Retrieval](https://mlsanity.com/paper/2410.09713/) (2024)
- [Stepwise Guided Policy Optimization: Coloring your Incorrect Reasoning in GRPO](https://mlsanity.com/paper/2505.11595/) (2025)
- [Auto-Formulating Dynamic Programming Problems with Large Language Models](https://mlsanity.com/paper/2507.11737/) (2025)
- [BPP-Search: Enhancing Tree of Thought Reasoning for Mathematical Modeling Problem Solving](https://mlsanity.com/paper/2411.17404/) (2024)

## Built on by (most cited first, 26 of 26)

- [A Survey of Process Reward Models: From Outcome Signals to Process Supervisions for Large Language Models](https://mlsanity.com/paper/2510.08049/) (2025)
- [Enhancing Large Language Model Reasoning with Reward Models: An Analytical Survey](https://mlsanity.com/paper/2510.01925/) (2025)
- [FrontierOR: Benchmarking LLMs' Capacity for Efficient Algorithm Design in Large-Scale Optimization](https://mlsanity.com/paper/2605.25246/) (2026)
- [ToolPRM: Fine-Grained Inference Scaling of Structured Outputs for Function Calling](https://mlsanity.com/paper/2510.14703/) (2025)
- [Strategy-Aware Optimization Modeling with Reasoning LLMs](https://mlsanity.com/paper/2605.02545/) (2026)
- [From Soliloquy to Agora: Memory-Enhanced LLM Agents with Decentralized Debate for Optimization Modeling](https://mlsanity.com/paper/2604.25847/) (2026)
- [MIRROR: A Multi-Agent Framework with Iterative Adaptive Revision and Hierarchical Retrieval for Optimization Modeling in Operations Research](https://mlsanity.com/paper/2602.03318/) (2026)
- [Execution-Verified Reinforcement Learning for Optimization Modeling](https://mlsanity.com/paper/2604.00442/) (2026)
- [OR-Space: A Full-Lifecycle Workspace Benchmark for Industrial Optimization Agents](https://mlsanity.com/paper/2605.28158/) (2026)
- [Beyond Objective Equivalence: Constraint Injection for LLM-Based Optimization Modeling on Vehicle Routing Problems](https://mlsanity.com/paper/2606.04816/) (2026)
- [OmniHarness: Harnessing Generalizable Visual Generation via Symbolic Policy Learning](https://mlsanity.com/paper/2609.16057/) (2026)
- [Automated Reformulation of Robust Optimization via Memory-Augmented Large Language Models](https://mlsanity.com/paper/2605.11813/) (2026)
- [MetaResearcher: Scaling Deep Research via Self-Reflective Reinforcement Learning in Adversarial Virtual Environments](https://mlsanity.com/paper/2606.19893/) (2026)
- [MiniOpt: Reasoning to Model and Solve General Optimization Problems with Limited Resources](https://mlsanity.com/paper/2606.25832/) (2026)
- [PEARL: Solver-in-the-Loop Interactive Optimization Modeling from Natural Language](https://mlsanity.com/paper/2607.18256/) (2026)
- [Co-evolving Agent Architectures and Interpretable Reasoning for Automated Optimization](https://mlsanity.com/paper/2604.17708/) (2026)
- [Training-Free Inference-Time Self-Reflection and Cost-Bounded Early Stopping for Large Language Models](https://mlsanity.com/paper/2608.18884/) (2026)
- [FormuEvo: LLM-Guided Evolution for Discovering Solver-Efficient Mixed-Integer Programming Formulations](https://mlsanity.com/paper/2608.23353/) (2026)
- [SOVER: Formal Certification of Optimization Reformulations via LLM-Assisted SMT Verification](https://mlsanity.com/paper/2609.00728/) (2026)
- [Ask Before You Optimize: Dynamic Pre-Formulation Clarification for Interactive Optimization](https://mlsanity.com/paper/2609.05258/) (2026)
- [Beyond Verified Answers: Solver-Informed Self-Distillation for Bootstrapping Operations Research Language Models](https://mlsanity.com/paper/2609.09957/) (2026)
- [STRETCH the Boundaries: A Unified Self-Taught Framework for Progressive LLM Evolution](https://mlsanity.com/paper/2609.18642/) (2026)
- [OptiSkill: A Hierarchical and Evolving SkillBank for LLM-Based Optimization Modeling](https://mlsanity.com/paper/2609.22987/) (2026)
- [Learning to Optimize through Solver-Grounded Self-Play](https://mlsanity.com/paper/2609.34205/) (2026)
- [LLMs as Adaptive Meta-Solvers: Strategy-Diverse RL for Industrial-Scale Optimization](https://mlsanity.com/paper/2609.34427/) (2026)
- [Right Answers, Costly Models: The Efficiency Gap in LLM-based Optimization Modeling](https://mlsanity.com/paper/2609.38884/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) have shown promising capabilities for solving Operations Research (OR) problems. While reinforcement learning serves as a powerful paradigm for LLM training on OR problems, existing works generally face two key limitations. First, outcome reward suffers from the credit assignment problem, where correct final answers can reinforce flawed reasoning. Second, conventional discriminative process supervision is myopic, failing to evaluate the interdependent steps of OR modeling holistically. To this end, we introduce StepORLM, a novel self-evolving framework with generative process supervision. At its core, StepORLM features a co-evolutionary loop where a policy model and a generative process reward model (GenPRM) iteratively improve on each other. This loop is driven by a dual-feedback mechanism: definitive, outcome-based verification from an external solver, and nuanced, holistic process evaluation from the GenPRM. The combined signal is used to align the policy via Weighted Direct Preference Optimization (W-DPO) and simultaneously refine the GenPRM. Our resulting 8B-parameter StepORLM establishes a new state-of-the-art across six benchmarks, significantly outperforming vastly larger generalist models, agentic methods, and specialized baselines. Moreover, the co-evolved GenPRM is able to act as a powerful and universally applicable process verifier, substantially boosting the inference scaling performance of both our own model and other existing LLMs.