| title | UniPath: Adaptive Coordination of Understanding and Generation for Unified Multimodal Reasoning |
|---|---|
| authors | Hayes Bai, Yinyi Luo, Wenwen Wang, Qingsong Wen, Jindong Wang |
| date | 2026-05-12 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2605.11400 |
| affiliation | The College of William & Mary |
| upvotes | 3 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf |
| abstract | Unified multimodal models (UMMs) aim to integrate understanding and generation within a single architecture. However, it remains underexplored how to effectively coordinate these two capabilities for more effective and efficient reasoning. Existing coordination approaches either perform coupling during training, without explicit inference-time coordination, or impose a fixed coordination pattern for all inputs. In this work, we show that multimodal tasks exhibit substantial coordination-path diversity: different inputs favor different coordination paths. This suggests that exploiting such diversity is key to improving performance. We propose UniPath, a framework for adaptively modeling and exploiting coordination-path diversity. Instead of enforcing a single coordination pattern, we represent task solving as the selection and execution of a path, ranging from direct answering to textual inference, visual-thought construction, and hypothesis-based exploration. We construct role-aligned trajectories to train a path-conditioned executor and introduce a lightweight planner mechanism to enable input-dependent path selection. Experiments show that leveraging coordination-path diversity improves performance over fixed coordination strategies while providing interpretable intermediate behaviors. The code is available at:https://github.com/AIFrontierLab/TorchUMM/tree/main/src/umm/post_training/unipath. |
cite
@misc{bai2026unipath,
title={UniPath: Adaptive Coordination of Understanding and Generation for Unified Multimodal Reasoning},
author={Hayes Bai and Yinyi Luo and Wenwen Wang and Qingsong Wen and Jindong Wang},
year={2026},
eprint={2605.11400},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2605.11400}
}