# RAVEN: Real-time Autoregressive Video Extrapolation with Consistency-model GRPO

Source: [RAVEN: Real-time Autoregressive Video Extrapolation with Consistency-model GRPO](<https://www.sota2.com/research/paper/6a06a8cbdbc836b87d82a399>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-18T06:00:01.259Z

Authors: Yanzuo Lu, Ronglai Zuo, Jiankang Deng

Published: 2026-05-14T17:59:30.000Z

## Abstract

Causal autoregressive video diffusion models support real-time streaming generation by extrapolating future chunks from previously generated content. Distilling such generators from high-fidelity bidirectional teachers yields competitive few-step models, yet a persistent gap between the history distributions encountered during training and those arising at inference constrains generation quality over long horizons. We introduce the Real-time Autoregressive Video Extrapolation Network (RAVEN), a training-time test framework that repacks each self rollout into an interleaved sequence of clean historical endpoints and noisy denoising states. This formulation aligns training attention with inference-time extrapolation and allows downstream chunk losses to supervise the history representations on which future predictions depend. We further propose Consistency-model Group Relative Policy Optimization (CM-GRPO), which reformulates a consistency sampling step as a conditional Gaussian transition and applies online Reinforcement Learning (RL) directly to this kernel, avoiding the Euler-Maruyama auxiliary process adopted in prior flow-model RL formulations. Experiments demonstrate that RAVEN surpasses recent causal video distillation baselines across quality, semantic, and dynamic degree evaluations, and that CM-GRPO provides further gains when combined with RAVEN.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Text-to-Video Generation on VBench (Total, Quality, and Detailed Temporal Metrics)](<https://www.sota2.com/research/sota/text-to-video-generation-on-vbench-total-quality-and-detailed-temporal-metrics.md>): Text-to-Video Generation; VBench; Total Score: 85.46.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2605.15190>)

[PDF](<https://arxiv.org/pdf/2605.15190.pdf>)

[Implementation](<https://github.com/mvp-ai-lab/RAVEN>)