5 papers
Ideas in Inference-time Scaling can Benefit Generative Pre-training Algorithms
Jiaming Song, Linqi Zhou
Generative pre-training is often framed through a false dichotomy between autoregressive models for discrete signals and diffusion models for continuous signals. We argue that the…
Terminal Velocity Matching
Linqi Zhou, Mathias Parger, Ayaan Haque +1
We propose Terminal Velocity Matching (TVM), a generalization of flow matching that enables high-fidelity one- and few-step generative modeling. TVM models the transition between a…
Inductive Moment Matching
Linqi Zhou, Stefano Ermon, Jiaming Song
Diffusion models and Flow Matching generate high-quality samples but are slow at inference, and distilling them into few-step models often leads to instability and extensive tuning…
3D-Adapter: Geometry-Consistent Multi-View Diffusion for High-Quality 3D Generation
Hansheng Chen, Bokui Shen, Yulin Liu +7
Multi-view image diffusion models have significantly advanced open-domain 3D object generation. However, most existing models rely on 2D network architectures that lack inherent 3D…
Personalized Preference Fine-tuning of Diffusion Models
Meihua Dang, Anikait Singh, Linqi Zhou +2
RLHF techniques like DPO can significantly improve the generation quality of text-to-image diffusion models. However, these methods optimize for a single reward that aligns model g…