AutoOR: Scalably Post-training LLMs to Autoformulate Operations Research Problems

[Submitted on 18 Apr 2026 (v1), last revised 20 Aug 2026 (this version, v4)]

View PDF HTML (experimental)

Abstract:Optimization problems are central to decision-making in manufacturing, logistics, scheduling, and other industrial settings. Translating complicated descriptions of these problems into solver-ready formulations requires specialized operations research (OR) expertise, making it hard to scale. We present AutoOR, a scalable synthetic data generation and reinforcement learning pipeline that trains LLMs to autoformulate optimization problems specified in natural language across linear, mixed-integer, and non-linear categories. AutoOR generates verified training data from standard optimization forms and uses solver execution feedback as the reward signal for RL post-training. AutoOR applied to an 8B model achieves state-of-the-art or competitive results across six established OR benchmarks, matching significantly larger frontier models. For a non-linear problem class involving physical dynamics, where frontier models score near 0%, we introduce a curriculum RL strategy that bootstraps from limited initial training data to make this class tractable for post-training. We believe that methods such as AutoOR can significantly accelerate industrial decision-making with AI.

Submission history

From: Weishi Yan [view email]
[v1] Sat, 18 Apr 2026 03:24:54 UTC (1,812 KB)
[v2] Wed, 6 May 2026 17:41:05 UTC (1,812 KB)
[v3] Wed, 19 Aug 2026 16:48:08 UTC (1,812 KB)
[v4] Thu, 20 Aug 2026 18:53:16 UTC (1,812 KB)