Search

huggingface.co › papers › 2606.01476

Paper page - OmniOPD: Logit-Free On-Policy Distillation via Speculative Verification

…Logit-Free On-Policy Distillation via Speculative Verification Published on May 31 Submitted by Yuhang Zhou on Jun 3 Meta Research Authors: , , , , , , , Abstract OmniOPD addresses limitations of standard On-Policy Distillation by…

Jun 3, 2026

huggingface.co › papers › 2605.12034

Paper page - Boosting Omni-Modal Language Models: Staged Post-Training with Visually Debiased Evaluation

…Staged Post-Training with Visually Debiased Evaluation Published on May 13 Submitted by liu on May 15 StepFun Authors: , , , , , , Abstract Research demonstrates that current omni-modal benchmarks may inflate performance through visual…

May 15, 2026