mlsanityData to 14 Oct 2026

Synthesizing Feature Extractors: An Agentic Approach for Algorithm Selection

arXiv 2608.17170 ↗First posted 17 Aug 2026cs.AIcited by 0

On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks

Abstract (arXiv)

Algorithm selection for constraint satisfaction problems requires extracting features that capture problem structure. Manually designing feature extractors demands deep domain expertise and quickly becomes a bottleneck when new problem classes appear. We present an automated approach that uses Large Language Models (LLMs) in an agentic check--fix--verify loop to synthesize executable Python scripts that act as interpretable, problem-specific feature extractors. Given a high-level MiniZinc model and an instance, the LLM agent generates code that constructs a typed graph representation and computes structural properties such as graph density, variable clustering, and constraint tightness. We evaluate our approach on three combinatorial problems (vehicle routing, car sequencing, fixed-length error-correcting codes) with a portfolio of five state-of-the-art solvers. The synthesized extractors yield algorithm selectors that consistently outperform both expert-curated mzn2feat features (up to 8.3 percentage points (pp) test-set accuracy on FLECC) and the best transformer-based trans2feat variants. In the meanwhile, the synthesized feature extractors remain inspectable.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 9 of its 31 references that are in the map. First posted on arXiv 17 Aug 2026.

Builds on

9 of 31 references.

References · 9

  1. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  2. 2020Auto-Sklearn 2.0: Hands-free AutoML via Meta-Learningcited by 174
  3. 2023Large Language Models for Automated Data Science: Introducing CAAFE for Context-Aware Automated Feature Engineeringcited by 127
  4. 2018Automated Algorithm Selection: Survey and Perspectivescited by 114
  5. 2024Large Language Models Can Automatically Engineer Features for Few-Shot Tabular Learningcited by 81
  6. 2023Large Language Model-Enhanced Algorithm Selection: Towards Comprehensive Algorithm Representationcited by 21
  7. 2013An Enhanced Features Extractor for a Portfolio of Constraint Solverscited by 5
  8. 2024GraSS: Combining Graph Neural Networks with Expert Knowledge for SAT Solver Selectioncited by 5
  9. 2013LLAMA: Leveraging Learning to Automatically Manage Algorithmscited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Improving Constraint Models with LLM Agents
  2. 2025HeuriGym: An Agentic Benchmark for LLM-Crafted Heuristics in Combinatorial Optimization
  3. 2025Behavior and Representation in Open-Weight Large Language Models for Combinatorial Optimization: From Feature Extraction to Algorithm Selection
  4. 2026DRAGON: LLM-Driven Decomposition and Reconstruction Agents for Large-Scale Combinatorial Optimization
  5. 2025CP-Agent: Agentic Constraint Programming
  6. 2026Agentic Proposing: Enhancing Large Language Model Reasoning via Compositional Skill Synthesis
  7. 2025CO-Bench: Benchmarking Language Model Agents in Algorithm Search for Combinatorial Optimization
  8. 2025ACCORD: Autoregressive Constraint-satisfying Generation for COmbinatorial Optimization with Routing and Dynamic attention