5 papers
Agentic ESOpt: Fine-Tuning Long-Horizon LLM Agents with Minimal GPU Requirements
Zhi Zheng, Rongsheng Chen, Yunpeng Ba +3
Reinforcement Learning (RL) has been promising in single-turn LLM fine-tuning. However, long-horizon agentic reasoning introduces increasingly branching interactions and sparse rew…
Hyper-ES: Effective Evolution Strategies for LLM Reasoning via Descent Direction Merging
Yu Gu, Zhi Zheng, Yunpeng Ba +3
Evolution Strategy (ES) is a promising alternative to gradient-based fine-tuning for resource-constrained Large Language Model (LLM) reasoning. However, directly applying ES to bil…
Rethinking Neural Combinatorial Optimization for Vehicle Routing Problems with Different Constraint Tightness Degrees
Fu Luo, Yaoxin Wu, Zhi Zheng +1
Recent neural combinatorial optimization (NCO) methods have shown promising problem-solving ability without requiring domain-specific expertise. Most existing NCO methods use train…
Monte Carlo Tree Search for Comprehensive Exploration in LLM-Based Automatic Heuristic Design
Zhi Zheng, Zhuoliang Xie, Zhenkun Wang +1
Handcrafting heuristics for solving complex optimization tasks (e.g., route planning and task allocation) is a common practice but requires extensive domain knowledge. Recently, La…
UDC: A Unified Neural Divide-and-Conquer Framework for Large-Scale Combinatorial Optimization Problems
Zhi Zheng, Changliang Zhou, Tong Xialiang +2
Single-stage neural combinatorial optimization solvers have achieved near-optimal results on various small-scale combinatorial optimization (CO) problems without requiring expert k…