Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation
Abstract
A novel Causal Forcing method addresses the architectural gap in distilling bidirectional video diffusion models into autoregressive models by using AR teachers for ODE initialization, significantly improving video generation performance.
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: https://thu-ml.github.io/CausalForcing.github.io/{https://thu-ml.github.io/CausalForcing.github.io/}
Community
Causal Forcing exposes a mathematical fallacy in Self Forcing and significantly outperforms it in both visual quality and motion dynamics, while maintaining the same training budget and inference efficiency, enabling real time streaming video generation on a single RTX 4090.
- Code (full-stack open source): https://github.com/thu-ml/Causal-Forcing
- Page: https://thu-ml.github.io/CausalForcing.github.io/
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- AutoRefiner: Improving Autoregressive Video Diffusion Models via Reflective Refinement Over the Stochastic Sampling Path (2025)
- LiveTalk: Real-Time Multimodal Interactive Video Diffusion via Improved On-Policy Distillation (2025)
- BAgger: Backwards Aggregation for Mitigating Drift in Autoregressive Video Diffusion Models (2025)
- Past- and Future-Informed KV Cache Policy with Salience Estimation in Autoregressive Video Diffusion (2026)
- Knot Forcing: Taming Autoregressive Video Diffusion Models for Real-time Infinite Interactive Portrait Animation (2025)
- Reward-Forcing: Autoregressive Video Generation with Reward Feedback (2026)
- End-to-End Training for Autoregressive Video Diffusion via Self-Resampling (2025)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment:
@librarian-bot
recommend
Models citing this paper 1
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper