Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation

• URL: http://arxiv.org/abs/2602.02214v2
• Date: Fri, 06 Feb 2026 04:48:17 GMT
• Title: Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation
• Abstract summary: Current methods distill pretrained bidirectional video diffusion models into few-step autoregressive (AR) models.<n>We propose Causal Forcing that uses an AR teacher for ODE.<n>Our method outperforms all baselines across all metrics.
• Score: 47.295642561210606
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: To achieve real-time interactive video generation, current methods distill pretrained bidirectional video diffusion models into few-step autoregressive (AR) models, facing an architectural gap when full attention is replaced by causal attention. However, existing approaches do not bridge this gap theoretically. They initialize the AR student via ODE distillation, which requires frame-level injectivity, where each noisy frame must map to a unique clean frame under the PF-ODE of an AR teacher. Distilling an AR student from a bidirectional teacher violates this condition, preventing recovery of the teacher's flow map and instead inducing a conditional-expectation solution, which degrades performance. To address this issue, we propose Causal Forcing that uses an AR teacher for ODE initialization, thereby bridging the architectural gap. Empirical results show that our method outperforms all baselines across all metrics, surpassing the SOTA Self Forcing by 19.3\% in Dynamic Degree, 8.7\% in VisionReward, and 16.7\% in Instruction Following. Project page and the code: \href{https://thu-ml.github.io/CausalForcing.github.io/}{https://thu-ml.github.io/CausalForcing.github.io/}

Related papers

• OPSD-V: On-Policy Self-Distillation for Post-Training Few-Step Autoregressive Video Generators [68.69611976783243]
  OPSD-V is an on-policy self-distillation paradigm for post-training few-step autoregressive (AR) video diffusion models.<n>Key idea is to introduce real long-video data as temporal context during training and use it to provide dense trajectory-level supervision.<n>Experiments show consistent improvements in visual quality, motion dynamics, and VBenchLong scores.
  arXiv  Detail & Related papers  (2026-07-09T17:59:11Z)
• Causal Forcing++: Scalable Few-Step Autoregressive Diffusion Distillation for Real-Time Interactive Video Generation [50.55853275866995]
  Real-time interactive video generation requires low-latency, streaming, and controllable rollout.<n>In this paper, we study a more aggressive setting: frame-wise autoregression with only 1--2 sampling steps.<n>We propose textbfCausal Forcing++, a principled and scalable pipeline.
  arXiv  Detail & Related papers  (2026-05-14T17:46:36Z)
• Skip to the Good Part: Representation Structure & Inference-Time Layer Skipping in Diffusion vs. Autoregressive LLMs [39.20818218608441]
  Autoregressive (AR) language models form representations incrementally through left-to-right prediction, whereas diffusion language models (dLLMs) are trained via full-sequence denoising.<n>We find that diffusion objectives result in different, more hierarchical abstractions with substantial early-layer redundancy and reduced recency bias.<n>We introduce a static, task-agnostic inference-time layer-skipping method requiring no architectural changes or KV-cache sharing.<n>Native dLLMs achieve up to 18.75% FLOPs reduction while preserving over 90% performance on reasoning and code generation benchmarks, whereas AR models degrade sharply under comparable skipping
  arXiv  Detail & Related papers  (2026-03-08T05:31:52Z)
• Probing to Refine: Reinforcement Distillation of LLMs via Explanatory Inversion [84.20493238687187]
  We introduce a novel framework that moves beyond simple mimicry to instill a deeper conceptual understanding.<n>underlinetextitFirst, to address pattern memorization, Explanatory Inversion (EI) generates targeted explanatory probes''<n>underlinetextitSecond, to improve generalization, Explanatory GRPO (texttEXGRPO) uses a reinforcement learning algorithm with a novel Dialogue Structure Utility Bonus.
  arXiv  Detail & Related papers  (2026-02-26T23:01:46Z)
• Light Forcing: Accelerating Autoregressive Video Diffusion via Sparse Attention [28.598033369607723]
  textscLight Forcing is a textitfirst sparse attention solution tailored for AR video generation models.<n>It incorporates a textitChunk-Aware Growth mechanism to quantitatively estimate the contribution of each chunk.<n>We also introduce a textit Sparse Attention to capture informative historical and local context in a coarse-to-fine manner.
  arXiv  Detail & Related papers  (2026-02-04T17:41:53Z)
• Self Forcing: Bridging the Train-Test Gap in Autoregressive Video Diffusion [67.94300151774085]
  We introduce Self Forcing, a novel training paradigm for autoregressive video diffusion models.<n>It addresses the longstanding issue of exposure bias, where models trained on ground-truth context must generate sequences conditioned on their own imperfect outputs.
  arXiv  Detail & Related papers  (2025-06-09T17:59:55Z)
• Single Trajectory Distillation for Accelerating Image and Video Style Transfer [22.304420035048942]
  Diffusion-based stylization methods typically denoise from a specific partial noise state for image-to-image and video-to-video tasks.<n>We propose single trajectory distillation (STD) starting from a specific partial noise state.<n>Our method surpasses existing acceleration models in terms of style similarity and aesthetic evaluations.
  arXiv  Detail & Related papers  (2024-12-25T16:40:23Z)
• From Slow Bidirectional to Fast Autoregressive Video Diffusion Models [48.35054927704544]
  Current video diffusion models achieve impressive generation quality but struggle in interactive applications due to bidirectional attention dependencies.<n>We address this limitation by adapting a pretrained bidirectional diffusion transformer to an autoregressive transformer that generates frames on-the-fly.<n>Our model achieves a total score of 84.27 on the VBench-Long benchmark, surpassing all previous video generation models.
  arXiv  Detail & Related papers  (2024-12-10T18:59:50Z)
• One-for-All: Bridge the Gap Between Heterogeneous Architectures in Knowledge Distillation [69.65734716679925]
  Knowledge distillation has proven to be a highly effective approach for enhancing model performance through a teacher-student training scheme. Most existing distillation methods are designed under the assumption that the teacher and student models belong to the same model family. We propose a simple yet effective one-for-all KD framework called OFA-KD, which significantly improves the distillation performance between heterogeneous architectures.
  arXiv  Detail & Related papers  (2023-10-30T11:13:02Z)
• Self-Distilled Masked Auto-Encoders are Efficient Video Anomaly Detectors [117.61449210940955]
  We propose an efficient abnormal event detection model based on a lightweight masked auto-encoder (AE) applied at the video frame level. We introduce an approach to weight tokens based on motion gradients, thus shifting the focus from the static background scene to the foreground objects. We generate synthetic abnormal events to augment the training videos, and task the masked AE model to jointly reconstruct the original frames.
  arXiv  Detail & Related papers  (2023-06-21T06:18:05Z)
• DisWOT: Student Architecture Search for Distillation WithOut Training [0.0]
  We explore a novel training-free framework to search for the best student architectures for a given teacher. Our work first empirically show that the optimal model under vanilla training cannot be the winner in distillation. Our experiments on CIFAR, ImageNet and NAS-Bench-201 demonstrate that our technique achieves state-of-the-art results on different search spaces.
  arXiv  Detail & Related papers  (2023-03-28T01:58:45Z)
• EmbedDistill: A Geometric Knowledge Distillation for Information Retrieval [83.79667141681418]
  Large neural models (such as Transformers) achieve state-of-the-art performance for information retrieval (IR) We propose a novel distillation approach that leverages the relative geometry among queries and documents learned by the large teacher model. We show that our approach successfully distills from both dual-encoder (DE) and cross-encoder (CE) teacher models to 1/10th size asymmetric students that can retain 95-97% of the teacher performance.
  arXiv  Detail & Related papers  (2023-01-27T22:04:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.