Everything is a Video: Unifying Modalities through Next-Frame Prediction

📅 2024-11-15
🏛️ arXiv.org
📈 Citations: 2
Influential: 0
📄 PDF

career value

174K/year
🤖 AI Summary
Traditional multimodal approaches rely on modality-specific encoders and late fusion, limiting scalability and cross-modal generalization. This work proposes a unified paradigm that reformulates diverse multimodal tasks—including text, image, audio, and video processing—as “next-frame prediction,” with all inputs and outputs represented as serialized video frames, enabling end-to-end, single-model inference. It pioneers the task-redefinition strategy—previously confined to NLP—within multimodal learning, eliminating modality-specific design in favor of a fully modality-agnostic architecture. The framework integrates cross-modal tokenization, sequential frame representation, autoregressive modeling, and a shared Transformer decoder, supporting joint multi-task pretraining. Empirical evaluation demonstrates strong zero-shot and few-shot generalization across text-to-text, image-to-text, video-to-video, video-to-text, and audio-to-text tasks. Crucially, adaptation requires only lightweight task-specific heads, underscoring its flexibility and efficiency.

Technology Category

Application Category

📝 Abstract
Multimodal learning, which involves integrating information from various modalities such as text, images, audio, and video, is pivotal for numerous complex tasks like visual question answering, cross-modal retrieval, and caption generation. Traditional approaches rely on modality-specific encoders and late fusion techniques, which can hinder scalability and flexibility when adapting to new tasks or modalities. To address these limitations, we introduce a novel framework that extends the concept of task reformulation beyond natural language processing (NLP) to multimodal learning. We propose to reformulate diverse multimodal tasks into a unified next-frame prediction problem, allowing a single model to handle different modalities without modality-specific components. This method treats all inputs and outputs as sequential frames in a video, enabling seamless integration of modalities and effective knowledge transfer across tasks. Our approach is evaluated on a range of tasks, including text-to-text, image-to-text, video-to-video, video-to-text, and audio-to-text, demonstrating the model's ability to generalize across modalities with minimal adaptation. We show that task reformulation can significantly simplify multimodal model design across various tasks, laying the groundwork for more generalized multimodal foundation models.
Problem

Research questions and friction points this paper is trying to address.

Unifying diverse modalities via next-frame prediction
Overcoming modality-specific encoder limitations in multimodal learning
Enabling seamless knowledge transfer across different tasks and modalities
Innovation

Methods, ideas, or system contributions that make the work stand out.

Unifies modalities via next-frame prediction
Treats inputs as sequential video frames
Simplifies model design across tasks
G
G. Thomas
Department of Computer Sciences, Durham University
H
Hudson Dean
Department of Computer Sciences, Durham University
S
Slack Thomas
Department of Computer Sciences, Durham University
W
Winterbottom Jamie
Department of Computer Sciences, Durham University
S
Sterling Chenghao
Department of Computer Sciences, Durham University
J
Junjie Xiao
Department of Computer Sciences, Durham University
S
Shentu Noura
Department of Computer Sciences, Durham University
A
Al Moubayed
Department of Computer Sciences, Durham University