Diffusion-based multimodal large language models (dMLLMs) decode by iterative denoising with bidirectional attention, which precludes the standard autoregressive key--value cache. Practical deployments therefore rely on approximate caching that reuses intermediate states across denoising steps. Such caching, however, o...
Qi-Yan Zhao, Xiao-Feng Zhang, Xu-Yao Zhang et al.· 0 citations
A training-free decoding method that scores each masked position by a multiplicative composite of softmax confidence and neighbor proximity, promoting contextually ready tokens above isolated candidates while suppressing low-confidence positional noise, operating block-free to keep high-readiness anchors globally acces...
Yi-Kai Zhao, Qiyan Zhao, Jia-Quan Zhang et al.· 0 citations
Systematic skill-bank audits reveal two recurring reliability failures: coalition pollution, where bank-level gains conceal negative coalition-level skill contributions, and cross-domain utility reversal, where source-beneficial skills reverse their effects after transfer.
Qiyan Zhao, Xiao-Feng Zhang, Bo Liu et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
