🏝️ On Vacation

Junxian Li

Duke-de-Artois

·

https://lijunxian111.github.io/

AI & ML interests

None yet

Recent Activity

upvoted a paper about 1 hour ago

Just MLPs: Efficient Visual State Reconstruction for Multimodal Language Models

upvoted a paper about 2 hours ago

Fewer Tokens, More Self-Teaching: On-Policy Self-Distillation for Extreme Visual Token Reduction

submitted a paper about 2 hours ago

Fewer Tokens, More Self-Teaching: On-Policy Self-Distillation for Extreme Visual Token Reduction

View all activity

Organizations

Duke-de-Artois 's papers 11

arxiv:2605.12357

arxiv:2603.08316

arxiv:2602.06663

arxiv:2510.14359

arxiv:2508.09456

arxiv:2508.08401

arxiv:2506.00189

arxiv:2412.19191

arxiv:2412.14642

arxiv:2411.18203

arxiv:2408.07246