paper-with-me

홈 › Papers

UGen: Unified Autoregressive Multimodal Model with Progressive Vocabulary Learning

2025-03-27 · Hongxuan Tang, Hao liu, Xinyan Xiao

We introduce UGen, a unified autoregressive multimodal model that demonstrates strong performance across text processing, image understanding, and image generation tasks simultaneously. UGen converts both texts and images into discrete token sequences and utilizes a single transformer to generate them uniformly in an autoregressive manner. To address the challenges associated with unified multimodal learning, UGen is trained using a novel mechanism, namely progressive vocabulary learning. In this process, visual token IDs are incrementally activated and integrated into the training phase, ultimately enhancing the effectiveness of unified multimodal learning. Experiments on comprehensive text and image tasks show that UGen achieves a significant overall performance improvement of 13.3% compared to the vanilla unified autoregressive method, and it also delivers competitive results across all tasks against several task-specific models.

📄 PDF Abstract BibTeX arXiv:2503.21193

Code (0)

등록된 구현이 없습니다.

Tasks

Image Generation

Similar Papers 제목 키워드 기반

STAR: STacked AutoRegressive Scheme for Unified Multimodal Learning

2025-12-15 · Jie Qin, Jiancheng Huang, Limeng Qiao, Lin Ma arxiv

Multimodal large language models (MLLMs) play a pivotal role in advancing the quest for general artificial intelligence. However, achieving unified target for multimodal understanding and generation remains challenging d…

Token by Token, Compromised: Backdoor Vulnerabilities in Unified Autoregressive Models

2026-05-19 · Tobias Braun, Jonas Henry Grebe, Hossein Shakibania, Anna Rohrbach 외 arxiv

Unified autoregressive models (UAMs) are transformer models that generate text as well as image tokens within a single autoregressive pass. Shared parameters and a multimodal vocabulary simplify the training pipeline and…

multimodal generationImage Generation

Unified Multimodal Autoregressive Modeling with Shared Context-Visual Tokenizer is Key to Unification

2026-06-16 · Wujian Peng, Lingchen Meng, Yuxuan Cai, Xianwei Zhuang 외 arxiv

Unified Multimodal Modeling aims to integrate visual understanding and generation within a single system. However, existing approaches typically rely on two disparate visual tokenizers, which splits the representation sp…

Reinforcement LearningImage GenerationImage Editing

MAUGen: A Unified Diffusion Approach for Multi-Identity Facial Expression and AU Label Generation

2026-01-31 · Xiangdong Li, Ye Lou, Ao Gao, Wei Zhang 외 arxiv

The lack of large-scale, demographically diverse face images with precise Action Unit (AU) occurrence and intensity annotations has long been recognized as a fundamental bottleneck in developing generalizable AU recognit…

Representation Learning

MUGEN: A Unified Framework for Efficient Motion Understanding and Generation

2026-07-30 · Zhankai Ye, Yukai Jin, Bingyang Wei, Bofan Li 외 arxiv

Grounding human motion in language, and language in motion, is a central step toward physical AI systems that can understand, generate, and communicate human behavior. Unified motion--language systems first coupled the t…