Efficient video generation requires reducing the quadratic cost of self-attention over long spatio-temporal token sequences. Existing efficient-attention methods typically apply the same computation pattern to every token, even though denoising difficulty varies substantially across video regions and evolves throughout...
Olga Zatsarynna, Denis Korzhenkov, Juergen Gall et al.· 0 citations
This work introduces PrismQuant, a quantizer-aware rotation framework that aligns the leading activation eigenspace with the constant group subspace of asymmetric grouped INT4, and formulate rotation design as a Ky Fan trace maximization and derive a closed-form solution that is provably optimal for this alignment obje...
Yan-Long Chen, Yi-Ning Chen, Song Zhang et al.· 0 citations
This work proposes SQuad, a Sub-Quadratic Attention Distillation framework that achieves a complexity of $\mathcal{O}(n\sqrt{n})$ in the resulting distilled Attention, naturally balancing the efficiency v/s expressivity trade-off.
Animesh Karnewar, Denis Korzhenkov, A. Habibian et al.· 0 citations
This work proposes Adaptive Vector-Quantized (AVQ) Attention, which adaptively allocates codebook capacity based on attention importance and develops an implementation using custom Triton kernels that enables the full adaptive refinement process, including importance scoring, child codeword insertion, and parent contri...
W.V.S.O. van den Dool, Patrick Forré, A. Habibian et al.· arXiv.org· 0 citations
UniCaMo is presented, a unified framework that enables simultaneous control of object trajectories and camera viewpoints by directly constructing the input noise of the diffusion model and achieves state-of-the-art results in both video quality and motion controllability on standard controllable video generation benchm...
Long Vu, T. Ngô, Animesh Karnewar et al.· 0 citations
MobileWan becomes the first 5B-scale video diffusion model deployable on a commercial mobile device and proposes a learnable attention head pruning method based on binary per-head gates optimized end-to-end using a noise-biased sparsity objective and distillation-based finetuning.
Mohsen Ghafoorian, Denis Korzhenkov, Adil Karjauv et al.· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
