Latent reward models (LRMs) enable efficient alignment of video diffusion models by scoring intermediate states directly in latent space. However, we find that optimizing against a fixed latent reward rapidly leads to latent reward hacking: the predicted reward stays high while perceptual and motion quality deteriorate...
Zhao-Long Su, Yu-Jin Han, Feng Wang et al.· 0 citations
Recent work incorporates reusable skills distilled from past interactions into multimodal agent training, providing procedural guidance for long-horizon planning and tool use. However, policy optimization in these methods remains driven primarily by sparse outcome rewards, providing little supervision for intermediate...
Bing Jiang, Guo-Xi Zhang, Jasper Wang et al.· 0 citations
On-policy distillation (OPD) has emerged as a core component of modern LLM post-training pipelines, yet we reveal a failure mode: degenerate agreement, where students exploit repetitive loops to achieve near-perfect token agreement with the teacher despite globally flawed responses. We therefore shift our focus from ag...
Zichao Yu, Chengzhi Yu, Shengze Xu et al.· 1 citation
This work proposes ReBind, a systematic framework that introduces semantic instructions with embedded reference tokens as the intermediate representation for multi-reference image-conditioned video editing and develops ReBind-Instruct, a specialized MLLM that learns to establish explicit bindings between visual attribu...
Xin-Yu Liu, Shi-Hao Li, Weihong Lin et al.· arXiv.org· 2 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
