Product
Help
Pricing
About
Jobs
Log in
Get started
Author
Add to collection
Junkang Wu
Observed name:
Wu, Junkang
Institution:
University of Science and Technology of China
Observed institution:
University of Science and Technology of China
OpenAlex ID:
A5151846710
Know When to Stop, Where to Restart: Accelerating Multi-Turn Agentic On-Policy Distillation
2026
·
Gui, Zhiyu
, Huang, Kexin
, et al.
·
arXiv (Cornell University)
PDF
PEA-DPO: Perception-Enhanced Alignment Direct Preference Optimization for MLLMs Alignment
2026
·
Feng, Jiawei
, Wu, Jiancan
, et al.
·
arXiv (Cornell University)
PDF
ARMOR: Stabilizing On-Policy LLM RL with Off-Policy Anchor Samples
2026
·
Huang, Kexin
, Wu, Junkang
, et al.
·
arXiv (Cornell University)
PDF
Experience Augmented Policy Optimization for LLM Reasoning
2026
·
Lu, Jinda
, Huang, Kexin
, et al.
·
arXiv (Cornell University)
PDF
R^2-Mem: Reflective Experience for Memory Search
2026
·
Wang, Xinyuan
, Mao, Wenyu
, et al.
·
arXiv (Cornell University)
PDF
AdViP: Aligning Multi-modal LLMs via Adaptive Vision-enhanced Preference Optimization
2026
·
Jinda Lu
, Jinghan Li
, et al.
·
International Journal of Computer Vision
Bridging Perception and Reasoning: Token Reweighting for RLVR in Multimodal LLMs
2026
·
Lu, Jinda
, Wu, Junkang
, et al.
·
arXiv (Cornell University)
PDF
On the Direction of RLVR Updates for LLM Reasoning: Identification and Exploitation
2026
·
Huang, Kexin
, Meng, Haoming
, et al.
·
arXiv (Cornell University)
PDF
Beyond Where to Look: Trajectory-Guided Reinforcement Learning for Multimodal RLVR
2026
·
Jinda Lu
, Junkang Wu
, et al.
·
Lecture notes in computer science
PDF
Show more works
View as search
Works count:
9
Citation count:
H-index:
I10-index:
Year
Topic
Type
