Image-Free Timestep Distillation via Continuous-Time Consistency with Trajectory-Sampled Pairs
Abstract
TBCM, a self-contained trajectory-based distillation method, enhances diffusion model efficiency by eliminating external data dependency and improving knowledge transfer, achieving high-quality generation with reduced computational resources.
Timestep distillation is an effective approach for improving the generation efficiency of diffusion models. The Consistency Model (CM), as a trajectory-based framework, demonstrates significant potential due to its strong theoretical foundation and high-quality few-step generation. Nevertheless, current continuous-time consistency distillation methods still rely heavily on training data and computational resources, hindering their deployment in resource-constrained scenarios and limiting their scalability to diverse domains. To address this issue, we propose Trajectory-Backward Consistency Model (TBCM), which eliminates the dependence on external training data by extracting latent representations directly from the teacher model's generation trajectory. Unlike conventional methods that require VAE encoding and large-scale datasets, our self-contained distillation paradigm significantly improves both efficiency and simplicity. Moreover, the trajectory-extracted samples naturally bridge the distribution gap between training and inference, thereby enabling more effective knowledge transfer. Empirically, TBCM achieves 6.52 FID and 28.08 CLIP scores on MJHQ-30k under one-step generation, while reducing training time by approximately 40% compared to Sana-Sprint and saving a substantial amount of GPU memory, demonstrating superior efficiency without sacrificing quality. We further reveal the diffusion-generation space discrepancy in continuous-time consistency distillation and analyze how sampling strategies affect distillation performance, offering insights for future distillation research. GitHub Link: https://github.com/hustvl/TBCM.
Community
Efficient Timestep Distillation Without Real Images.
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
- From Structure to Detail: Hierarchical Distillation for Efficient Diffusion Model (2025)
- Score Distillation of Flow Matching Models (2025)
- Large Scale Diffusion Distillation via Score-Regularized Continuous-Time Consistency (2025)
- Towards One-step Causal Video Generation via Adversarial Self-Distillation (2025)
- Advancing End-to-End Pixel Space Generative Modeling via Self-supervised Pre-training (2025)
- DDTime: Dataset Distillation with Spectral Alignment and Information Bottleneck for Time-Series Forecasting (2025)
- Self-Forcing++: Towards Minute-Scale High-Quality Video Generation (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 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper