Structure-Aware Skill Retrieval and Arbitration for Optimization Modeling
cs.AI
Submitted: 2026-10-30
Updated: 2026-10-30
License: http://creativecommons.org/licenses/by/4.0/
The gist: Large Language Models (LLMs) can turn natural-language optimization tasks into solver-checked answers…
Terminology
Abstract
Large Language Models (LLMs) can turn natural-language optimization tasks into solver-checked answers, but current agents remain brittle under narrative rewrites and often store reusable experience at the instance level. We introduce ArchOR, an archetype-centric skill agent that builds on OptSkills by adding two retrieval-focused modules. Centroid-Aware Representation Learning (CARL) refines the ordering of cached skill representations through a low-rank prototype residual. Signature-Guided Retrieval (S R) attaches open-dimensional structural signatures to skills, prioritizes candidates by weighted Jaccard overlap, and bounds semantic re-ranking to a top- K set. A baseline-preserving arbitration rule determines when the structural route should replace the original selector. With the same DeepSeek-V4.1-Flash backbone and released skill library, ArchOR reaches 65.00% Micro-Avg. Pass@1 across five benchmarks, improving over OptSkills by 1.64 pp Micro-Avg. and 2.68 pp Macro-Avg.
Sources
- EvoSkill: Automated Skill Discovery for Multi-Agent Systems
- DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
- XSkill: Continual Learning from Experience and Skills in Multimodal Agents
- SkillRet: A Large-Scale Benchmark for Skill Retrieval in LLM Agents
- AlphaOPT: Formulating Optimization Programs with Self-Improving LLM Experience Library
- Large-Scale Optimization Model Auto-Formulation: Harnessing LLM Flexibility via Structured Workflow
- AutoOR: Scalably Post-training LLMs to Autoformalize Operations Research Problems
- Trace2Skill: Distill Trajectory-Local Lessons into Transferable Agent Skills
- MIRROR: A Multi-Agent Framework with Iterative Adaptive Revision and Hierarchical Retrieval for Optimization Modeling in Operations Research
- Qwen3 Technical Report
- ORThought: Benchmarking and Automating Logistics Optimization Modeling
- OptSkills: Learning Generalizable Optimization Skills from Problem Archetypes via Cluster-Based Distillation
- AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution
- MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents
- OptiSkill: A Hierarchical and Evolving SkillBank for LLM-Based Optimization Modeling
- Memento-Skills: Let Agents Design Agents
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents