{"ID":5937754,"CreatedAt":"2026-07-07T03:14:33.014478982Z","UpdatedAt":"2026-07-08T17:29:38.726141074Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.04423","arxiv_id":"2607.04423","title":"Transferability Between Understanding and Generation in Unified Multimodal Models","abstract":"Unified Multimodal Models (UMMs) integrate image understanding and generation within a single architecture, yet how the two tasks interact remains understudied. We investigate $\\boldsymbol{\\mathsf{transferability}}$ in UMMs: whether training a capability on one task improves the same capability on the other without explicit supervision. Through controlled experiments, we empirically find that transferability depends on architecture-models with fully shared transformer backbone and a unified visual encoder exhibit consistent cross-task transfer, while loosely coupled designs show little or none. Leveraging this transferability, we propose a practical training strategy. The most straightforward way to improve a target generative capability (e.g., counting) is to fine-tune generation directly, but this can degrade visual quality due to distribution shift. Instead, we train the corresponding understanding task and let it transfer into generation, which improves capability-specific generative performance while minimizing distribution shift. We validate this across three capabilities-counting, spatial relation, and text recognition/generation-showing that cross-task transferability can be systematically exploited in UMMs.","short_abstract":"Unified Multimodal Models (UMMs) integrate image understanding and generation within a single architecture, yet how the two tasks interact remains understudied. We investigate $\\boldsymbol{\\mathsf{transferability}}$ in UMMs: whether training a capability on one task improves the same capability on the other without exp...","url_abs":"https://arxiv.org/abs/2607.04423","url_pdf":"https://arxiv.org/pdf/2607.04423v1","authors":"[\"Jiwon Kang\",\"Heeji Yoon\",\"Jaewoo Jung\",\"Jaewon Min\",\"Minkyeong Jeon\",\"Biyeon Hwang\",\"Sangwon Jung\",\"Seungryong Kim\"]","published":"2026-07-05T17:33:59Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[\"Transformer\"]","has_code":false}