Showing cs.AIShow all
2 papers · 1 filter
cs.AI2026
HorizonFlow: Variable-Length Planning for Offline Goal-Conditioned RL
JunHyeok Oh, Zian Jang, Byung-Jun Lee
Recent advances in generative planning have made trajectory inpainting a promising approach to offline goal-conditioned reinforcement learning. However, these methods typically spe…
cs.AI2025
Rethinking DPO: The Role of Rejected Responses in Preference Misalignment
Jay Hyeon Cho, JunHyeok Oh, Myunsoo Kim +1
Direct Preference Optimization (DPO) is a simple and efficient framework that has attracted substantial attention. However, it often struggles to meet its primary objectives -- inc…