UniEvo-VL: An On-policy Self-Distillation Training Recipe for Multimodal Model Self-improvement
Paper • 2609.38721 • Published • 275
Andrew Lee
Andrewleewyn
·
AI & ML interests
mlops, question answering
Recent Activity
upvoted a paper 4 days ago
MaLiang-Harness: A Programmable Path to Image and Video GenerationOrganizations
None yet
upvoted 4
papers 4 days ago
MaLiang-Harness: A Programmable Path to Image and Video Generation
Paper • 2609.34309 • Published • 409
False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents
Paper • 2609.39102 • Published • 392
The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation
Paper • 2609.36484 • Published • 465