Unison: Benchmarking Unified Multimodal Models via Synergistic Understanding and Generation

Jinyu Liu, Xincheng Shuai, Henghui Ding, Yu-Gang Jiang
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:78454-78464, 2026.

Abstract

Unified multimodal models capable of both understanding and generation have achieved remarkable strides. However, despite their unified designs, existing evaluations typically assess understanding and generation capabilities in isolation, overlooking the synergy between comprehension and generation. To bridge this gap, we introduce Unison, a comprehensive benchmark comprising 2,169 high-quality unified task samples, designed to evaluate joint understanding and generation in unified multimodal models. Unison offers three key strengths: 1) Comprehensive Dimensions: Unison encompasses internal consistency, understanding-guided generation, generation-guided understanding, and mutual enhancement to enable holistic evaluation. 2) Diagnostic Evaluation: it provides both unified and decoupled tracks for understanding and generation, allowing fine-grained attribution of failure modes and quantitative analysis of the gains from unified modeling. 3) Human Alignment: we also train Unison-Judge, an evaluation model well aligned with human judgments to achieve reliable assessment. Based on systematic evaluations of state-of-the-art models on Unison, we uncover critical limitations in current unified multimodal systems and highlight promising directions for future research. Unison will be publicly available at https://github.com/FudanCVL/Unison.

Cite this Paper


BibTeX
@InProceedings{pmlr-v306-liu26dv, title = {Unison: Benchmarking Unified Multimodal Models via Synergistic Understanding and Generation}, author = {Liu, Jinyu and Shuai, Xincheng and Ding, Henghui and Jiang, Yu-Gang}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {78454--78464}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/liu26dv/liu26dv.pdf}, url = {https://proceedings.mlr.press/v306/liu26dv.html}, abstract = {Unified multimodal models capable of both understanding and generation have achieved remarkable strides. However, despite their unified designs, existing evaluations typically assess understanding and generation capabilities in isolation, overlooking the synergy between comprehension and generation. To bridge this gap, we introduce Unison, a comprehensive benchmark comprising 2,169 high-quality unified task samples, designed to evaluate joint understanding and generation in unified multimodal models. Unison offers three key strengths: 1) Comprehensive Dimensions: Unison encompasses internal consistency, understanding-guided generation, generation-guided understanding, and mutual enhancement to enable holistic evaluation. 2) Diagnostic Evaluation: it provides both unified and decoupled tracks for understanding and generation, allowing fine-grained attribution of failure modes and quantitative analysis of the gains from unified modeling. 3) Human Alignment: we also train Unison-Judge, an evaluation model well aligned with human judgments to achieve reliable assessment. Based on systematic evaluations of state-of-the-art models on Unison, we uncover critical limitations in current unified multimodal systems and highlight promising directions for future research. Unison will be publicly available at https://github.com/FudanCVL/Unison.} }
Endnote
%0 Conference Paper %T Unison: Benchmarking Unified Multimodal Models via Synergistic Understanding and Generation %A Jinyu Liu %A Xincheng Shuai %A Henghui Ding %A Yu-Gang Jiang %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-liu26dv %I PMLR %P 78454--78464 %U https://proceedings.mlr.press/v306/liu26dv.html %V 306 %X Unified multimodal models capable of both understanding and generation have achieved remarkable strides. However, despite their unified designs, existing evaluations typically assess understanding and generation capabilities in isolation, overlooking the synergy between comprehension and generation. To bridge this gap, we introduce Unison, a comprehensive benchmark comprising 2,169 high-quality unified task samples, designed to evaluate joint understanding and generation in unified multimodal models. Unison offers three key strengths: 1) Comprehensive Dimensions: Unison encompasses internal consistency, understanding-guided generation, generation-guided understanding, and mutual enhancement to enable holistic evaluation. 2) Diagnostic Evaluation: it provides both unified and decoupled tracks for understanding and generation, allowing fine-grained attribution of failure modes and quantitative analysis of the gains from unified modeling. 3) Human Alignment: we also train Unison-Judge, an evaluation model well aligned with human judgments to achieve reliable assessment. Based on systematic evaluations of state-of-the-art models on Unison, we uncover critical limitations in current unified multimodal systems and highlight promising directions for future research. Unison will be publicly available at https://github.com/FudanCVL/Unison.
APA
Liu, J., Shuai, X., Ding, H. & Jiang, Y.. (2026). Unison: Benchmarking Unified Multimodal Models via Synergistic Understanding and Generation. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:78454-78464 Available from https://proceedings.mlr.press/v306/liu26dv.html.

Related Material