analiza aiqbench
This paper introduces EPOCH, a novel framework for LLM-based automated heuristic design that integrates instance coverage objectives with experiential preference optimization to improve combinatorial problem solving. By enabling the generator to internalize reflection feedback through direct preference optimization and reflective self-distillation, it achieves superior performance over standard fine-tuning and inference-only baselines across various problem scales. The method represents a significant step forward in leveraging LLMs for complex optimization tasks without freezing the model or relying solely on scalar fitness signals.
EPOCH: Coverage-Aware Evolution with Experiential Preference Optimization for LLM-based Automated Heuristic Design
arXiv:2610.17379v1 Announce Type: new Abstract: LLM-based automated heuristic design (AHD) discovers heuristics for combinatorial optimization problems (COPs), where an LLM generates candidates in an evolutionary search. Existing frameworks freeze the generator or fine-tune it on a scalar fitness signal hiding which instances each heuristic covers. Reflection-based frameworks improve heuristics with natural-language feedback, but it stays in the prompt and never reaches the generator's weights. We propose EPOCH, which makes instance coverage the objective of the evolutionary search and fine-tunes the generator to internalize reflection. Parent selection, operator allocation, and stagnation recovery all optimize how well the heuristics jointly cover the training instances, so the search returns a complementary heuristic ensemble. The generator learns from its own search by experiential preference optimization. A teacher LLM returns a natural-language insight on the generated heuristics, and the generator produces new heuristics with this insight. These improvements join the preference pairs for direct preference optimization (DPO). Since DPO only teaches the generator to rank these improvements, a reflective self-distillation (RSD) term makes it more likely to generate them on its own, and the gains accumulate over the long horizon of the search. Across COPs and open-source LLM backbones, EPOCH's heuristic ensembles outperform inference-only and fine-tuned baselines and generalize across problem scales.
Abstract:LLM-based automated heuristic design (AHD) discovers heuristics for combinatorial optimization problems (COPs), where an LLM generates candidates in an evolutionary search. Existing frameworks freeze the generator or fine-tune it on a scalar fitness signal hiding which instances each heuristic covers. Reflection-based frameworks improve heuristics with natural-language feedback, but it stays in the prompt and never reaches the generator's weights. We propose EPOCH, which makes instance coverage the objective of the evolutionary search and fine-tunes the generator to internalize reflection. Parent selection, operator allocation, and stagnation recovery all optimize how well the heuristics jointly cover the training instances, so the search returns a complementary heuristic ensemble. The generator learns from its own search by experiential preference optimization. A teacher LLM returns a natural-language insight on the generated heuristics, and the generator produces new heuristics with this insight. These improvements join the preference pairs for direct preference optimization (DPO). Since DPO only teaches the generator to rank these improvements, a reflective self-distillation (RSD) term makes it more likely to generate them on its own, and the gains accumulate over the long horizon of the search. Across COPs and open-source LLM backbones, EPOCH's heuristic ensembles outperform inference-only and fine-tuned baselines and generalize across problem scales.
Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
arXiv:2610.17379 [cs.AI]
(or arXiv:2610.17379v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2610.17379 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 300 cuvinte
Subiecte conexe
[AINews] High Taste Testers Ep. 1
Latent Space · 17 oct. 2026
Our response to Séb Krier on Plan A
LessWrong · 14 iul. 2026
Dharma vs Adharma - The Perplexity of Being Alive
hillpost.in · 25 aug. 2026
[Paper] Distillation for Incrimination and Distillation for Capabilities
LessWrong · 9 oct. 2026