paper-with-me

홈 › Papers

An Empirical Study of Autoregressive Pre-training from Videos

2025-01-09 · Jathushan Rajasegaran, Ilija Radosavovic, Rahul Ravishankar, Yossi Gandelsman, Christoph Feichtenhofer, Jitendra Malik

We empirically study autoregressive pre-training from videos. To perform our study, we construct a series of autoregressive video models, called Toto. We treat videos as sequences of visual tokens and train transformer models to autoregressively predict future tokens. Our models are pre-trained on a diverse dataset of videos and images comprising over 1 trillion visual tokens. We explore different architectural, training, and inference design choices. We evaluate the learned visual representations on a range of downstream tasks including image recognition, video classification, object tracking, and robotics. Our results demonstrate that, despite minimal inductive biases, autoregressive pre-training leads to competitive performance across all benchmarks. Finally, we find that scaling our video models results in similar scaling curves to those seen in language models, albeit with a different rate. More details at https://brjathu.github.io/toto/

📄 PDF Abstract BibTeX arXiv:2501.05453

Code (0)

등록된 구현이 없습니다.

Tasks

Object TrackingVideo Classification

Similar Papers 제목 키워드 기반

Accelerating Training of Autoregressive Video Generation Models via Local Optimization with Representation Continuity

2026-04-08 · Yucheng Zhou, Jianbing Shen arxiv

Autoregressive models have shown superior performance and efficiency in image generation, but remain constrained by high computational costs and prolonged training times in video generation. In this study, we explore met…

Video GenerationImage Generation

Loong: Generating Minute-level Long Videos with Autoregressive Language Models

2024-10-03 · Yuqing Wang, Tianwei Xiong, Daquan Zhou, Zhijie Lin 외

It is desirable but challenging to generate content-rich long videos in the scale of minutes. Autoregressive large language models (LLMs) have achieved great success in generating coherent and long sequences of tokens in…

Video Generation

Forcing-KV: Hybrid KV Cache Compression for Efficient Autoregressive Video Diffusion Models

2026-05-10 · Yicheng Ji, Zhizhou Zhong, Jun Zhang, Qin Yang 외 arxiv

Autoregressive (AR) video diffusion models adopt a streaming generation framework, enabling long-horizon video generation with real-time responsiveness, as exemplified by the Self Forcing training paradigm. However, exis…

Video Generation

HARP: Autoregressive Latent Video Prediction with High-Fidelity Image Generator

2022-09-15 · Younggyo Seo, Kimin Lee, Fangchen Liu, Stephen James 외

Video prediction is an important yet challenging problem; burdened with the tasks of generating future frames and learning environment dynamics. Recently, autoregressive latent video models have proved to be a powerful v…

Data AugmentationPredictionVideo PredictionVocal Bursts Intensity Prediction

Autoregressivity in the Latent Space of a GP-VAE Language Model: An Empirical Ablation Study

2025-12-30 · Yves Ruffenach arxiv

This paper provides an ablation-based analysis of latent autoregression in GP-VAE models, building upon our previous work introducing the architecture. Language models typically rely on an autoregressive factorization ov…