| title | A^2RD: Agentic Autoregressive Diffusion for Long Video Consistency |
|---|---|
| authors | Do Xuan Long, Yale Song, Min-Yen Kan, Tomas Pfister, Long T. Le |
| date | 2026-05-07 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2605.06924 |
| affiliation | |
| upvotes | 15 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf |
| abstract | Synthesizing consistent and coherent long video remains a fundamental challenge. Existing methods suffer from semantic drift and narrative collapse over long horizons. We present A^2RD, an Agentic Auto-Regressive Diffusion architecture that decouples creative synthesis from consistency enforcement. A^2RD formulates long video synthesis as a closed-loop process that synthesizes and self-improves video segment-by-segment through a Retrieve--Synthesize--Refine--Update cycle. It comprises three core components: (i) Multimodal Video Memory that tracks video progression across modalities; (ii) Adaptive Segment Generation that switches among generation modes for natural progression and visual consistency; and (iii) Hierarchical Test-Time Self-Improvement that self-improves each segment at frame and video levels to prevent error propagation. We further introduce LVBench-C, a challenging benchmark with non-linear entity and environment transitions to stress-test long-horizon consistency. Across public and LVBench-C benchmarks spanning one- to ten-minute videos, A^2RD outperforms state-of-the-art baselines by up to 30% in consistency and 20% in narrative coherence. Human evaluations corroborate these gains while also highlighting notable improvements in motion and transition smoothness. |
cite
@misc{long2026rd,
title={A^2RD: Agentic Autoregressive Diffusion for Long Video Consistency},
author={Do Xuan Long and Yale Song and Min-Yen Kan and Tomas Pfister and Long T. Le},
year={2026},
eprint={2605.06924},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2605.06924}
}