Lijun Yu
I am a research scientist at Google DeepMind. I obtained my Ph.D. and M.S. at Carnegie Mellon University School of Computer Science. I graduated summa cum laude from Peking University major in Computer Science as well as Economics. My research interests lie around multi-modal foundation models, especially for video generation.
Research Areas
Authored Publications
Sort By
VideoPoet: A Large Language Model for Zero-Shot Video Generation
Dan Kondratyuk
Xiuye Gu
Grant Schindler
Rachel Hornung
Jimmy Yan
Ming-Chang Chiu
Hassan Akbari
Josh Dillon
Agrim Gupta
Meera Hahn
Anja Hauth
David Hendon
Alonso Martinez
Kihyuk Sohn
Xuan Yang
Huisheng Wang
Lu Jiang
ICML(2024)
Preview abstract
We present VideoPoet, a language model capable of synthesizing high-quality video, with matching audio, from a large variety of conditioning signals. VideoPoet employs a decoder-only transformer architecture that processes multimodal inputs -- including images, videos, text, and audio. The training protocol follows that of Large Language Models (LLMs), consisting of two stages: pretraining and task-specific adaptation. During pretraining, VideoPoet incorporates a mixture of multimodal generative objectives within an autoregressive Transformer framework. The pretrained LLM serves as a foundation that can be adapted for a range of video generation tasks. We present empirical results demonstrating the model's state-of-the-art capabilities in zero-shot video generation, specifically highlighting VideoPoet's ability to generate high-fidelity motions. Project page: http://sites.research.google/videopoet/
View details
Photorealistic Video Generation with Diffusion Models
Agrim Gupta
Kihyuk Sohn
Xiuye Gu
Fei-Fei Li
Lu Jiang
ECCV(2024)
Preview abstract
We present W.A.L.T, a transformer-based approach for photorealistic video generation via diffusion modeling. Our approach has two key design decisions. First, we use a causal encoder to jointly compress images and videos within a unified latent space, enabling training and generation across modalities. Second, for memory and training efficiency, we use a window attention architecture tailored for joint spatial and spatiotemporal generative modeling. Taken together these design decisions enable us to achieve state-of-the-art performance on established video (UCF-101 and Kinetics-600) and image (ImageNet) generation benchmarks without using classifier free guidance. Finally, we also train a cascade of three models for the task of text-to-video generation consisting of a base latent video diffusion model, and two video super-resolution diffusion models to generate videos of 512*896 resolution at 8 frames per second.
View details
Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation
Nitesh Bharadwaj Gundavarapu
Luca Versari
Kihyuk Sohn
Agrim Gupta
Xiuye Gu
Alex Hauptmann
Boqing Gong
Lu Jiang
ICLR(2024)
Preview abstract
While Large Language Models (LLMs) are the dominant models for generative tasks in language, they do not perform as well as diffusion models on image and video generation. To effectively use LLMs for visual generation, one crucial component is the visual tokenizer that maps pixel-space inputs to discrete tokens appropriate for LLM learning. In this paper, we introduce MAGVIT-v2, a video tokenizer designed to generate concise and expressive tokens for both videos and images using a common token vocabulary. Equipped with this new tokenizer, we show that LLMs outperform diffusion models on standard image and video generation benchmarks including ImageNet and Kinetics. In addition, we demonstrate that our tokenizer surpasses the previously top-performing video tokenizer on two more tasks: (1) video compression comparable to the next-generation video codec (VCC) according to human evaluations, and (2) learning effective representations for action recognition tasks.
View details
Preview abstract
Document understanding tasks, in particular, Visually-rich Document Entity Retrieval (VDER), have gained significant attention in recent years thanks to their broad applications in enterprise AI. However, publicly available data have been scarce for these tasks due to strict privacy constraints and high annotation costs. To make things worse, the non-overlapping entity spaces from different datasets hinder the knowledge transfer between document types. In this paper, we propose a method to collect massive-scale and weakly labeled data from the web to benefit the training of VDER models. The collected dataset, named DocumentNet, does not depend on specific document types or entity sets, making it universally applicable to all VDER tasks. The current DocumentNet consists of 30M documents spanning nearly 400 document types organized in a four-level ontology. Experiments on a set of broadly adopted VDER tasks show significant improvements when DocumentNet is incorporated into the pre-training for both classic and few-shot learning settings. With the recent emergence of large language models (LLMs), DocumentNet provides a large data source to extend their multimodal capabilities for VDER.
View details
SPAE: Semantic Pyramid AutoEncoder for Multimodal Generation with Frozen LLMs
Zhiruo Wang
Yonatan Bisk
Alex Hauptmann
Lu Jiang
NeurIPS(2023)
Preview abstract
In this work, we introduce Semantic Pyramid AutoEncoder (SPAE) for enabling frozen LLMs to perform both understanding and generation tasks involving non-linguistic modalities such as images or videos. SPAE converts between raw pixels and interpretable lexical tokens (or words) extracted from the LLM's vocabulary. The resulting tokens capture both the semantic meaning and the fine-grained details needed for visual reconstruction, effectively translating the visual content into a language comprehensible to the LLM, and empowering it to perform a wide array of multimodal tasks. Our approach is validated through in-context learning experiments with frozen PaLM 2 and GPT 3.5 on a diverse set of image understanding and generation tasks. Our method marks the first successful attempt to enable a frozen LLM to generate image content while surpassing state-of-the-art performance in image understanding tasks, under the same setting, by over 25%.
View details
MAGVIT: Masked Generative Video Transformer
Kihyuk Sohn
Han Zhang
Huiwen Chang
Alex Hauptmann
Lu Jiang
CVPR(2023)
Preview abstract
This paper introduces a Masked Generative Video Transformer, named MAGVIT, for multi-task video generation. We train a single MAGVIT model and apply it to multiple video generation tasks at inference time. To this end, two new designs are proposed: an improved 3D tokenizer model to quantize a video into spatial-temporal visual tokens, and a novel technique to embed conditions inside the mask to facilitate multi-task training.
We conduct extensive experiments to demonstrate the compelling quality, efficiency, and flexibility of the proposed model.
First, MAGVIT radically improves the previous best fidelity on two video generation tasks.
In terms of efficiency, MAGVIT offers leading video generation speed at inference time, which is estimated to be one or two orders-of-magnitudes faster than other models. As for flexibility, we verified that a single trained MAGVIT is able to generically perform 8+ tasks at several video benchmarks from drastically different visual domains. We will open source our framework and models.
View details
Preview abstract
Score-based modeling through stochastic differential equations (SDEs) has provided a new perspective on diffusion models, and demonstrated superior performance on continuous data. However, the gradient of the log-likelihood function, i.e., the score function, is not properly defined for discrete spaces. This makes it non-trivial to adapt the score-based modeling to categorical data. In this paper, we extend diffusion models to discrete variables by introducing a stochastic jump process where the reverse process denoises via a continuous-time Markov chain. This formulation admits an analytical simulation during backward sampling. To learn the reverse process, we extend score matching to general categorical data, and show that an unbiased estimator can be obtained via simple matching of the conditional marginal distributions. We demonstrate the effectiveness of the proposed method on a set of synthetic and real-world music and image benchmarks.
View details
