Showing cs.AIShow all
2 papers · 1 filter
cs.AI2026
Beyond Scripted Search: Sample-Efficient Reward Discovery via Agentic Black-box Optimization
Minghao Li, Rui Tan, Ruihang Wang
Designing dense reward functions for low-level reinforcement learning (RL) control remains difficult. Recent work uses large language models (LLMs) to iteratively generate and refi…
cs.AI2025
Phythesis: Physics-Guided Evolutionary Scene Synthesis for Energy-Efficient Data Center Design via LLMs
Minghao LI, Ruihang Wang, Rui Tan +1
Data center (DC) infrastructure serves as the backbone to support the escalating demand for computing capacity. Traditional design methodologies that blend human expertise with spe…