Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note2 min read

RAISE: LLM-based Automated Heuristic Design with Robust Adversary Instance Search

Source

arXiv

arxiv.org/abs/2606.31801v1

Read the source

Source

Plain-Language Brief

The paper proposes a framework called Robust Adversary Instance Search (RAISE), which uses Large Language Models (LLMs) to design heuristics that perform well under real-world distributional shifts. The framework optimizes the heuristics within a neighborhood of the training distribution, using an LLM-free inner loop to find difficult instances around the training set.

What the Document Says

  • The paper discusses Automated Heuristic Design (AHD) with Large Language Models (LLMs).
  • Existing LLM-based AHD methods optimize heuristics for a fixed training instance set but may fail under real-world distributional shifts.
  • The proposed framework, RAISE, integrates constrained worst-case instance search within the LLM-based evolutionary search loop.
  • RAISE treats robust AHD as a constrained adversarial instance search problem.
  • The outer loop evolves heuristics via LLM operators, while an LLM-free inner loop identifies hard instances around the training set.
  • The inner loop uses a basis distribution parameterization with boundary projection.
  • Experiments on Online Bin Packing (OBP), Online Job Shop Scheduling (OJSP), and Online Vehicle Routing (OVRP) demonstrate that existing LLM-based AHD methods degrade under distribution shift, but RAISE maintains strong performance across all tested distributions and problem scales.

Key Concepts Mentioned

  • Automated Heuristic Design
  • Large Language Models
  • Robust Adversary Instance Search
  • distributional shifts

Claims Made by the Document

  • Existing LLM-based AHD methods may fail under real-world distributional shifts.
  • RAISE maintains strong performance across all tested distributions and problem scales.

Evidence Used

  • Comprehensive experiments on OBP, OJSP, and OVRP across five distribution families

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Automated Heuristic Design (AHD) with Large Language Models (LLMs) has shown remarkable progress in discovering high-quality heuristics. However, existing LLM-based AHD methods optimize heuristics for a fixed training instance set and may fail catastrophically when deployed under real-world distributional shifts. We propose Robust Adversary Instance Search (RAISE), a framework that integrates constrained worst-case instance search within a principled neighborhood of the training distribution into the LLM-based evolutionary search loop. RAISE treats robust AHD as a constrained adversarial instance search problem: the outer loop evolves heuristics via LLM operators, while an LLM-free inner loop efficiently identifies hard instances within an epsilon-ball around the training instance set using a basis distribution parameterization with boundary projection. Comprehensive experiments on Online Bin Packing (OBP), Online Job Shop Scheduling (OJSP), and Online Vehicle Routing (OVRP) across five distribution families demonstrate that existing LLM-based AHD methods degrade by up to 19 times under distribution shift, while RAISE consistently maintains strong performance across all tested distributions and problem scales