arXiv AI
Oct 5

TACD: Distilling Efficient Text-to-Motion Models via Terminal Amplification Control

The paper introduces TACD, a Terminal‑Amplification‑Controlled Distillation method that trains efficient text‑to‑motion generators using on‑policy flow distillation from pretrained teachers, without requiring real‑motion data. TACD addresses a failure mode where velocity matching over‑weights errors near the denoising endpoint by tying the latest teacher query to the student’s step size, thereby bounding loss weights in clean‑motion space. Experiments on HumanML3D, KIT‑ML, HY‑Motion, and Kimodo show that TACD improves few‑step generation quality, reduces FID scores, and achieves significant speedups and memory savings compared to teacher models.

By Wei-Jin Huang, Yuan-Ming Li, Kun-Yu Lin, Wang Luo, Yinlin Zhu, Yue Yu, Shenghao Ye, Junbin Yuan, Fa-Ting Hong, Qing Zhang, Wei-Shi Zheng
arXiv Computer Vision
Nov 5

VA-TDM: A Unified View for Improving Few-Step Diffusion Models by Velocity Adversarial Distillation

The paper introduces Velocity Adversarial Distillation (VAD), a method that uses a learned critic to compare velocity fields of student and teacher diffusion models, turning disagreements into learning signals. VAD unifies several existing distillation techniques and leads to VA‑TDM, which applies trajectory distribution matching and transition‑aware online velocity learning to supervise partial denoising along the student’s sampling path. Experiments demonstrate that VA‑TDM achieves state‑of‑the‑art performance for four‑step text‑to‑image and text‑to‑video generation, outperforming prior few‑step methods and approaching the quality of many‑step baselines.

By Yihong Luo, Tianyu Pang, Tianyang Hu
arXiv AI
Aug 18

Step-Level On-Policy Distillation: Interpolating Between On-Policy Distillation and Supervised Fine-Tuning

arXiv:2608. 16333v1 Announce Type: cross Abstract: On-policy distillation (OPD) aligns a student model with a teacher's logit distribution on student-generated trajectories.

By Changhui Sun, Lanbo Liu, Hang Lei, Tong Ling, Jiahang Xie, Zhiyong Zheng, Yujia Wang, Hao Liu, Feng Xiao, Lu Liu, Yanlong Du, Zifeng Cheng, Ziwei Jiang, Qing Gu
arXiv AI
Jun 10

Fast and Highly Expressive Policy Learning for Offline Reinforcement Learning via Bootstrapped Flow Q-Learning

arXiv:2606. 10613v1 Announce Type: cross Abstract: Diffusion-based Q-learning has emerged as a powerful paradigm for offline reinforcement learning, but its reliance on multi-step denoising makes both training and inference computationally expensive and brittle.

By Thanh Nguyen, Tri Ton, Hongbin Choe, Tung M. Luu, Chang D. Yoo
arXiv AI
Oct 5

Geometry-Aware Time Reparameterization for Flow-Map Distillation

The paper proposes a geometry‑aware time reparameterization for flow‑map distillation, which reallocates student time to trajectory segments with high normal acceleration while preserving the teacher’s geometric paths and terminal distribution. By deriving a shared clock that equalizes a normal‑acceleration statistic and incorporating it into Lagrangian flow‑map distillation, the method improves sample quality on synthetic data, CIFAR‑10, and CelebA‑64 compared to identity‑time distillation, even in one‑step generation scenarios.

By F\'elix Dedek, Makoto Yamada