Robust Heuristic Algorithm Design with LLMs
Not on the map. Why?
Abstract (arXiv)
We posit that we can generate more robust and performant heuristics if we augment approaches using LLMs for heuristic design with tools that explain why heuristics underperform and suggestions about how to fix them. We find even simple ideas that (1) expose the LLM to instances where the heuristic underperforms; (2) explain why they occur; and (3) specialize design to regions in the input space, can produce more robust algorithms compared to existing techniques~ -- ~the heuristics we produce have a ~28× better worst-case performance compared to FunSearch, improve average performance, and maintain the runtime.
Key facts. Cited by 3 papers in the map (data through 13 Oct 2026). Builds on 19 of its 66 references that are in the map. First posted on arXiv 9 Oct 2025.
Builds on
19 of 66 references.
References · 19
- 2021Evaluating Large Language Models Trained on Codecited by 8,205
- 2021Program Synthesis with Large Language Modelscited by 3,573
- 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,412
- 2022Competition-Level Code Generation with AlphaCodecited by 1,445
- 2024Long-context LLMs Struggle with Long In-context Learningcited by 286
- 2020DreamCoder: Growing generalizable, interpretable knowledge with wake-sleep Bayesian program learningcited by 161
- 2019Themis: Fair and Efficient GPU Cluster Schedulingcited by 43
- 2021TACCL: Guiding Collective Algorithm Synthesis using Communication Sketchescited by 38
- 2024AlphaVerus: Bootstrapping Formally Verified Code Generation through Self-Improving Translation and Treefinementcited by 33
- 2020Synthesizing Optimal Collective Algorithmscited by 30
- 2023What do LLMs need to Synthesize Correct Router Configurations?cited by 29
- 2023CASSINI: Network-Aware Job Scheduling in Machine Learning Clusterscited by 26
7 more
- 2019GAP: Generalizable Approximate Graph Partitioning Frameworkcited by 18
- 2021Solving Large-Scale Granular Resource Allocation Problems Efficiently with POPcited by 13
- 2024Systematic Evaluation of Long-Context LLMs on Financial Conceptscited by 6
- 2022Efficient Direct-Connect Topologies for Collective Communicationscited by 5
- 2023Solving Max-Min Fair Resource Allocations Quickly on Large Graphscited by 3
- 2023Finding Adversarial Inputs for Heuristics using Multi-level Optimizationcited by 3
- 2023Everything Matters in Programmable Packet Schedulingcited by 1
Built on by
Citing papers per year: 2025: 1; 2026: 2.
3 papers.