AI Paper Atlas

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

Strategy-Aware Optimization Modeling with Reasoning LLMs

Ruiqing Zhao, Fengzhi Li, Yuan Zuo, Rui Liu, YanSong Liu, Yunfei Ma, Fanyu Meng, Junlan Feng

ICML 2026 · Large language models › Agents, code and math · approach: LLM-guided program synthesis · 0 citations (OpenAlex)

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 RL with solver feedback after SFT (ablation)SAGE · 8-benchmark averagepass@1: 74.0 -> 80.3Improved
add structured reasoning template in RL (ablation)SAGE · 8-benchmark averagepass@1: 78.3 -> 80.3Improved
add structured reasoning template in RL (ablation)SAGE · MAMO Complexpass@1: 77.5 -> 84.7Improved
add segment-weighted GRPO (vs uniform weights) (ablation)SAGE · 8-benchmark averagepass@1: 76.7 -> 80.3Improved
add segment-weighted GRPO (vs uniform weights) (ablation)SAGE · MAMO Complexpass@1: 73.9 -> 84.7Improved
add efficiency reward (ablation)SAGE · 8-benchmark averagepass@1: 77.4 -> 80.3Improved
add efficiency reward (ablation)SAGE · MAMO Easypass@1: 95.0 -> 94.7No clear effect
add structured template in RL (ablation)SAGE · IndustryOR Hard subsetaccuracy: 33.3 -> 50.0Improved
add structured template in RL (ablation)SAGE · IndustryOR Easy subsetaccuracy: 81.8 -> 77.3Hurt
use SAGE-DS-14B instead of SIRL-Q2.5-7BSAGE · 8-benchmark averagepass@1: 72.7 -> 80.3Improved
use SAGE-Qwen3-8B instead of SIRL-Q2.5-7BSAGE with Qwen3-8B backbone · 8-benchmark averagepass@1: 72.7 -> 77.0Improved
use SAGE instead of SIRL for generation diversitySAGE · Pass@16 variable typesdistinct variable types: 1.80 -> 2.33Improved

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