Mode Seeking meets Mean Seeking for Fast Long Video Generation

TL;DR

This paper introduces a novel training paradigm combining Mode Seeking and Mean Seeking for efficient long video generation, leveraging a Decoupled Diffusion Transformer to improve coherence and fidelity over extended durations.

Contribution

It proposes a unified framework that decouples local fidelity from long-term coherence using a dual-head approach with supervised flow matching and mode-seeking divergence.

Findings

Effective long-range coherence in minute-scale videos

Improved local sharpness and motion realism

Significant reduction in generation time

Abstract

Scaling video generation from seconds to minutes faces a critical bottleneck: while short-video data is abundant and high-fidelity, coherent long-form data is scarce and limited to narrow domains. To address this, we propose a training paradigm where Mode Seeking meets Mean Seeking, decoupling local fidelity from long-term coherence based on a unified representation via a Decoupled Diffusion Transformer. Our approach utilizes a global Flow Matching head trained via supervised learning on long videos to capture narrative structure, while simultaneously employing a local Distribution Matching head that aligns sliding windows to a frozen short-video teacher via a mode-seeking reverse-KL divergence. This strategy enables the synthesis of minute-scale videos that learns long-range coherence and motions from limited long videos via supervised flow matching, while inheriting local realism by aligning every sliding-window segment of the student to a frozen short-video teacher, resulting in a few-step fast long video generator. Evaluations show that our method effectively closes the fidelity-horizon gap by jointly improving local sharpness, motion and long-range consistency. Project website: https://primecai.github.io/mmm/.