Pan Wang

wangpan-ustc

·

wangpan-ustc

AI & ML interests

LLM/VLM agent

Recent Activity

upvoted a paper about 3 hours ago

ProAS: A Progress-Aware Advantage Shaping Framework for Multi-Turn Agentic Reinforcement Learning

upvoted a paper about 3 hours ago

ChronoReplay: One Answer Interface Is Not Enough for Streaming Spatial Intelligence

authored a paper about 20 hours ago

Trajectory2Gym: Decision-Grounded Compositional Interaction Synthesis for Tool-Use Agents

View all activity

Organizations

None yet

Papers 8

arxiv:2610.33568

arxiv:2610.27628

arxiv:2610.20424

arxiv:2609.05571

models 1

wangpan-ustc/AtlasVA

Image-Text-to-Text • Updated May 27 • 7

datasets 0

None public yet