UniFork: Exploring Modality Alignment for Unified Multimodal Understanding and Generation
Abstract
A Y-shaped architecture, UniFork, balances shared learning and task specialization for unified image understanding and generation, outperforming conventional fully shared Transformer models.
Unified image understanding and generation has emerged as a promising paradigm in multimodal artificial intelligence. Despite recent progress, the optimal architectural design for such unified models remains an open challenge. In this work, we start by analyzing the modality alignment behaviors of task-specific expert models for understanding and generation, as well as current unified models. Our analysis reveals a crucial observation: understanding tasks benefit from a progressively increasing modality alignment across network depth, which helps build up semantic information for better comprehension; In contrast, generation tasks follow a different trend: modality alignment increases in the early layers but decreases in the deep layers to recover spatial details. These divergent alignment patterns create a fundamental conflict in fully shared Transformer backbones, where a uniform representational flow often leads to performance compromises across two tasks. Motivated by this finding, we introduce UniFork, a novel Y-shaped architecture that shares the shallow layers for cross-task representation learning, while employing task-specific branches in deeper layers to avoid task interference. This design effectively balances shared learning and task specialization. Through extensive ablation experiments, we demonstrate that Unifork consistently outperforms conventional fully shared Transformer architectures, and achieves performance on par with or better than task-specific models.
Community
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- Resolving Task Objective Conflicts in Unified Multimodal Understanding and Generation via Task-Aware Mixture-of-Experts (2025)
- Pisces: An Auto-regressive Foundation Model for Image Understanding and Generation (2025)
- Are Unified Vision-Language Models Necessary: Generalization Across Understanding and Generation (2025)
- Emerging Properties in Unified Multimodal Pretraining (2025)
- Nexus-Gen: A Unified Model for Image Understanding, Generation, and Editing (2025)
- HaploOmni: Unified Single Transformer for Multimodal Video Understanding and Generation (2025)
- Mogao: An Omni Foundation Model for Interleaved Multi-Modal Generation (2025)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment:
@librarian-bot
recommend
Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper