paper-with-me

홈 › Papers

Self-Supervised Flow Matching for Scalable Multi-Modal Synthesis

2026-03-06 · Hila Chefer, Patrick Esser, Dominik Lorenz, Dustin Podell, Vikash Raja, Vinh Tong, Antonio Torralba, Robin Rombach arxiv

Strong semantic representations improve the convergence and generation quality of diffusion and flow models. Existing approaches largely rely on external models, which require separate training, operate on misaligned objectives, and exhibit unexpected scaling behavior. We argue that this dependence arises from the model's training objective, which poses a denoising task with little incentive to learn semantic representations. We introduce Self-Flow: a self-supervised flow matching paradigm that integrates representation learning within the generative framework. Our key mechanism, Dual-Timestep Scheduling, applies heterogeneous noise levels across tokens, creating an information asymmetry that forces the model to infer missing information from corrupted inputs. This drives learning strong representations alongside generative capabilities without external supervision. Our method generalizes across modalities and enables multi-modal training while following expected scaling laws, achieving superior image, video, and audio generation.

📄 PDF Abstract BibTeX arXiv:2603.06507

Code (0)

등록된 구현이 없습니다.

Tasks

Representation LearningAudio Generation

Similar Papers 제목 키워드 기반

Self-Point-Flow: Self-Supervised Scene Flow Estimation from Point Clouds with Optimal Transport and Random Walk

2021-05-18 · CVPR 2021 1 · Ruibo Li, Guosheng Lin, Lihua Xie

Due to the scarcity of annotated scene flow data, self-supervised scene flow learning in point clouds has attracted increasing attention. In the self-supervised manner, establishing correspondences between two point clou…

Scene Flow EstimationSelf-Supervised LearningSelf-supervised Scene Flow Estimation

Flow2Stereo: Effective Self-Supervised Learning of Optical Flow and Stereo Matching

2020-04-05 · CVPR 2020 6 · Pengpeng Liu, Irwin King, Michael Lyu, Jia Xu

In this paper, we propose a unified method to jointly learn optical flow and stereo matching. Our first intuition is stereo matching can be modeled as a special case of optical flow, and we can leverage 3D geometry behin…

3D geometryOptical Flow EstimationSelf-Supervised LearningStereo Matching

Tilt Matching for Scalable Sampling and Fine-Tuning

2025-12-26 · Peter Potaptchik, Cheuk-Kit Lee, Michael S. Albergo arxiv

We propose a simple, scalable algorithm for using stochastic interpolants to sample from unnormalized densities and for fine-tuning generative models. The approach, Tilt Matching, arises from a dynamical equation relatin…

Vector-Symbolic Architecture for Event-Based Optical Flow

2024-05-14 · Hongzhi You, Yijun Cao, Wei Yuan, Fanjun Wang 외

From a perspective of feature matching, optical flow estimation for event cameras involves identifying event correspondences by comparing feature similarity across accompanying event frames. In this work, we introduces a…

Event-based Optical FlowOptical Flow EstimationSelf-Supervised Learning

Self-Assessed Generation: Trustworthy Label Generation for Optical Flow and Stereo Matching in Real-world

2024-10-14 · Han Ling, Yinghui Sun, Quansen Sun, Ivor Tsang 외

A significant challenge facing current optical flow and stereo methods is the difficulty in generalizing them well to the real world. This is mainly due to the high costs required to produce datasets, and the limitations…

Optical Flow EstimationStereo Matching