AI Paper Atlas

Research map / Large language models / Instruction and fine-tuning

Refining Hybrid Genetic Search for CVRP via Reinforcement Learning-Finetuned LLM

Rongjie Zhu, Cong Zhang, Zhiguang Cao

ICLR 2026 · Large language models › Instruction and fine-tuning · approach: Fine-tuning generalization and delta compression

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
fine-tune 14B LLM with RL to generate HGS crossover operator, vs expert operator (800 iters)HGS-PyVRP · CVRPLIB X n in [200,400)gap %: 1.85 -> 1.67Improved
RL-tuned operator on larger unseen sizes (800 iters)HGS-PyVRP · CVRPLIB X n in [800,1000]gap %: 2.32 -> 2.24Improved
RL-tuned operator on smallest instances (800 iters)HGS-PyVRP · CVRPLIB X n in [100,200)gap %: 0.62 -> 0.70Hurt
RL-tuned operator at 1000 iterationsHGS-PyVRP · CVRPLIB X n in [600,800)gap %: 2.43 -> 2.35Improved
prompt GPT-4o or GPT-o3 to modify the crossover operator instead of RL fine-tuningHGS-PyVRP · CVRPLIB Xgap %: 0.62 -> 0.62 (GPT-4o_800 n 100-200)No clear effect
use continuous relative-improvement reward instead of discrete +1 reward (ablation)RFTHGS · CVRPLIB X n in [200,400)gap %: 1.78 -> 1.67Improved
use continuous reward instead of discrete reward (ablation)RFTHGS · CVRPLIB X n in [800,1000]gap %: 2.30 -> 2.24Improved
RL fine-tune the 14B model instead of prompting GPT models for compilable operatorsoperator generation · 16 sampled operatorssuccessful compilation rate: 3/16 -> 16/16Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy