Skip to main content
Search

University of Science and Technology of China

3 related articles

Subscribe to updates
  1. arXiv

    Joint audio-video generation stops letting audio drag: OmniOPD steadies distillation by giving the audio teacher a cleaner video condition

    In joint audio-video generation, audio has far fewer tokens than video, so its supervision is sparse and distillation is unstable, and simply raising the audio loss weight makes things worse. OmniOPD has the audio teacher produce targets from a cleaner visual condition, samples several trajectories per prompt, and weights each timestep by teacher-student transition divergence; on JavisBench it raises video quality VQ from 2.714 for LTX-2.3 to 3.347 while holding audio quality AQ at 5.449, below the audio teacher's 5.576.

  2. arXiv

    Video flow alignment drops the either-or: guidance stays in sampling, and CFG-free inference keeps the reward gains

    Decoupling guided sampling from the optimization anchor lets a video flow model keep the gains of guided exploration under CFG-free single-branch inference while preserving the original semantics of inference-time guidance.

  3. arXiv

    Robot world models stop learning only from success: training on human egocentric video and perturbed failures aligns predicted success with real deployment

    The action-conditioned world model HF-WM trains on egocentric human manipulation video together with perturbed failure actions, reaching state-of-the-art action-conditioned video generation and predicted success rates that track real-world deployment.

You’re all caught up