REVIEW 3 cited by
MALT Diffusion: Memory-Augmented Latent Transformers for Any-Length Video Generation
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
MALT Diffusion: Memory-Augmented Latent Transformers for Any-Length Video Generation
read the original abstract
Diffusion models are successful for synthesizing high-quality videos but are limited to generating short clips (e.g., 2-10 seconds). Synthesizing sustained footage (e.g. over minutes) still remains an open research question. In this paper, we propose MALT Diffusion (using Memory-Augmented Latent Transformers), a new diffusion model specialized for long video generation. MALT Diffusion (or just MALT) handles long videos by subdividing them into short segments and doing segment-level autoregressive generation. To achieve this, we first propose recurrent attention layers that encode multiple segments into a compact memory latent vector; by maintaining this memory vector over time, MALT is able to condition on it and continuously generate new footage based on a long temporal context. We also present several training techniques that enable the model to generate frames over a long horizon with consistent quality and minimal degradation. We validate the effectiveness of MALT through experiments on long video benchmarks. We first perform extensive analysis of MALT in long-contextual understanding capability and stability using popular long video benchmarks. For example, MALT achieves an FVD score of 220.4 on 128-frame video generation on UCF-101, outperforming the previous state-of-the-art of 648.4. Finally, we explore MALT's capabilities in a text-to-video generation setting and show that it can produce long videos compared with recent techniques for long text-to-video generation.
Forward citations
Cited by 3 Pith papers
-
CineWeaver: Training-Free Reference-Controllable Multi-Shot Long Video Generation for Cinematic Storytelling
Inference-time manipulation of RoPE, attention masks, per-shot conditioning, and VAE decoding lets frozen text-to-video models produce reference-controlled multi-shot long videos.
-
TivTok: Broadcasting Time-Invariant Tokens for Scalable Video Tokenization
TivTok factorizes video clips into reusable time-invariant tokens and frame-specific time-variant tokens via Scope-Induced Factorization and Invariant Broadcasting, achieving 2.91x better compression for 128-frame vid...
-
EverAnimate: Minute-Scale Human Animation via Latent Flow Restoration
EverAnimate restores drifted latent flow trajectories in chunked video generation via persistent latent propagation and restorative flow matching, achieving measurable gains in PSNR, SSIM, LPIPS, and FID over prior lo...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.