We propose UniDFlow, a unified discrete flow-matching framework for multimodal understanding, generation, and editing. It decouples understanding and generation via task-specific low-rank adapters, avoiding objective interference and representation entanglement, while a novel reference-based multimodal preference alignment optimizes relative outcomes under identical conditioning, improving faithfulness and controllability without large-scale retraining. UniDFlpw achieves SOTA performance across eight benchmarks and exhibits strong zero-shot generalization to tasks including inpainting, in-context image generation, reference-based editing, and compositional generation, despite no explicit task-specific training.
@article{arxiv.2602.12221,
title = {Best of Both Worlds: Multimodal Reasoning and Generation via Unified Discrete Flow Matching},
author = {Onkar Susladkar and Tushar Prakash and Gayatri Deshmukh and Kiet A. Nguyen and Jiaxun Zhang and Adheesh Juvekar and Tianshu Bao and Lin Chai and Sparsh Mittal and Inderjit S Dhillon and Ismini Lourentzou},
journal= {arXiv preprint arXiv:2602.12221},
year = {2026}
}