Reinforcing Few-step Generators via Reward-Tilted Distribution Matching
Paper • 2605.26108 • Published • 7
AI & ML interests
None defined yet.
Recent Activity
View all activity
Papers
Beyond Uniform Token-Level Trust Region in LLM Reinforcement Learning
Flow-DPPO: Divergence Proximal Policy Optimization for Flow Matching Models
cheese1
authored 3
papers 17 days ago
cheese1
submitted a
paper to Daily Papers 19 days ago
Jiaqi-hkust
authored 2
papers 30 days ago
Jiaqi-hkust
posted an update 30 days ago
Post
3976
🚀 Introducing Robust-U1: Teaching MLLMs to Self-Recover Corrupted Visual Content
Multimodal Large Language Models (MLLMs) have achieved impressive visual understanding, yet they remain highly brittle under real-world corruptions—noise, blur, compression artifacts, adverse weather.
Standard MLLMs suffer dramatic performance drops, and existing robustness solutions come with fundamental limits: black‑box feature alignment lacks interpretability, while white‑box text reasoning cannot restore the lost pixel‑level visual details. This raises a crucial question:
🧐 Can MLLMs recover corrupted visual content by themselves?
If the answer is yes, we can move beyond merely “compensating” for corruption and instead build a more intrinsic, generalizable form of resilience. Robust-U1 is our answer to that question.
💡 Paper: https://arxiv.org/abs/2606.08063
🔗 Code: github.com/jqtangust/Robust-U1
🌍 Demo: Jiaqi-hkust/Robust-U1
Multimodal Large Language Models (MLLMs) have achieved impressive visual understanding, yet they remain highly brittle under real-world corruptions—noise, blur, compression artifacts, adverse weather.
Standard MLLMs suffer dramatic performance drops, and existing robustness solutions come with fundamental limits: black‑box feature alignment lacks interpretability, while white‑box text reasoning cannot restore the lost pixel‑level visual details. This raises a crucial question:
🧐 Can MLLMs recover corrupted visual content by themselves?
If the answer is yes, we can move beyond merely “compensating” for corruption and instead build a more intrinsic, generalizable form of resilience. Robust-U1 is our answer to that question.
💡 Paper: https://arxiv.org/abs/2606.08063
🔗 Code: github.com/jqtangust/Robust-U1
🌍 Demo: Jiaqi-hkust/Robust-U1
- 1 reply
Jiaqi-hkust
submitted a
paper to Daily Papers 30 days ago
cheese1
authored a
paper about 1 month ago
MeowFET
authored a
paper about 1 month ago
zhouxiangxin
authored a
paper about 1 month ago
P2333
authored a
paper about 1 month ago
zhouxiangxin
authored a
paper about 1 month ago
P2333
authored a
paper about 1 month ago
Jayce-Ping
authored a
paper about 1 month ago
zhouxiangxin
authored a
paper about 1 month ago
sumailmao
authored a
paper about 1 month ago
MeowFET
authored a
paper about 1 month ago
cheese1
authored 2
papers about 1 month ago