HiMPO: Hindsight-Informed Memory Policy Optimization for Less-Entangled Credit in Long-Horizon Agents
Paper • 2606.16285 • Published • 1
Jiangze Yan
Kadins
·
AI & ML interests
Main areas of interest include large language models and multimodal models.
Recent Activity
authored a paper 20 days ago
DAST: Difficulty-Adaptive Slow-Thinking for Large Reasoning Models authored a paper 20 days ago
HiMPO: Hindsight-Informed Memory Policy Optimization for Less-Entangled Credit in Long-Horizon Agents authored a paper 20 days ago
HEAL: Hindsight Entropy-Assisted Learning for Reasoning DistillationOrganizations
upvoted a paper 20 days ago