Back to AI News
arxiv

APEX Drops the Fake-Score Network for Few-Step Streaming Video Generation

Researchers extend APEX to streaming video generation by using condition shifting to estimate the velocity field of the generated distribution inside the generator, eliminating the separate fake-score network used in Distribution Matching Distillation. The original text condition drives generation while a shifted condition drives fake estimation with the same parameters, retaining S2-Avatar's frozen reference teacher, Self-Replay Forcing, and perceptual regularization. Evaluated on 180-second videos with a 70-case human study and a 320-case StreamAV-Bench comparison, APEX reports higher instruction fulfillment on the Progressive track and higher visual aesthetics across both tracks than S2-Avatar, while motion quality favors the baseline. Under matched training settings, APEX also reports 4.4% lower peak allocated GPU memory per GPU and 9.2% shorter mean iteration time than DMD.

AI Brief

It offers a concrete architectural simplification for cheaper few-step streaming video generation, backed by human and benchmark comparisons plus memory and iteration-time measurements.

arXiv:2611.14033v1 Announce Type: cross Abstract: Few-step sampling reduces the cost of streaming video generation. Distribution Matching Distillation (DMD) supervises few-step generators with a separate fake-score network, which adds training memory and computation and must track the evolving generator. We extend APEX to streaming video by using condition shifting to estimate the velocity field of the generated distribution within the generator. The original text condition selects generation, while a shifted condition selects fake estimation using the same parameters. This removes the independent fake network while retaining S2-Avatar's frozen reference teacher, Self-Replay Forcing, and perceptual regularization. We evaluate APEX on 180-second videos using three backbone evaluations per block and a one-step spatial Refiner, with a 70-case human study and a 320-case StreamAV-Bench comparison. Compared with S2-Avatar, APEX achieves higher video instruction fulfillment on the Progressive track (3.856 versus 3.831) and higher visual aesthetics across both tracks (0.641898 versus 0.632606). Human raters favor APEX most strongly on instruction following, with +30.0% net preference, followed by consistency and video quality at +4.3% each. Motion quality favors S2-Avatar, audiovisual synchronization is predominantly tied, and usability is unchanged. Separate measurements under matched training settings show 4.4% lower peak allocated GPU memory per GPU and 9.2% shorter mean iteration time than DMD.
Be the first to rate this article.

Comments & Questions (0)

Captcha
Please be respectful — let's keep the conversation friendly.

No comments yet

Be the first to comment!