UniGame: Turning a Unified Multimodal Model Into Its Own Adversary
Abstract
UniGame, a self-adversarial post-training framework, improves consistency, understanding, generation, and robustness in unified multimodal models by introducing a lightweight perturber at the shared token interface.
Unified Multimodal Models (UMMs) have shown impressive performance in both understanding and generation with a single architecture. However, UMMs still exhibit a fundamental inconsistency: understanding favors compact embeddings, whereas generation favors reconstruction-rich representations. This structural trade-off produces misaligned decision boundaries, degraded cross-modal coherence, and heightened vulnerability under distributional and adversarial shifts. In this paper, we present UniGame, a self-adversarial post-training framework that directly targets the inconsistencies. By applying a lightweight perturber at the shared token interface, UniGame enables the generation branch to actively seek and challenge fragile understanding, turning the model itself into its own adversary. Experiments demonstrate that UniGame significantly improves the consistency (+4.6%). Moreover, it also achieves substantial improvements in understanding (+3.6%), generation (+0.02), out-of-distribution and adversarial robustness (+4.8% and +6.2% on NaturalBench and AdVQA). The framework is architecture-agnostic, introduces less than 1% additional parameters, and is complementary to existing post-training methods. These results position adversarial self-play as a general and effective principle for enhancing the coherence, stability, and unified competence of future multimodal foundation models. The official code is available at: https://github.com/AIFrontierLab/UniGame
Community
Solving inconsistency issues of Unified Multimodal Models
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
- CoDefend: Cross-Modal Collaborative Defense via Diffusion Purification and Prompt Optimization (2025)
- Enhancing CLIP Robustness via Cross-Modality Alignment (2025)
- Multimodal Safety Is Asymmetric: Cross-Modal Exploits Unlock Black-Box MLLMs Jailbreaks (2025)
- JEPA-T: Joint-Embedding Predictive Architecture with Text Fusion for Image Generation (2025)
- ReMatch: Boosting Representation through Matching for Multimodal Retrieval (2025)
- Q-MLLM: Vector Quantization for Robust Multimodal Large Language Model Security (2025)
- When Alignment Fails: Multimodal Adversarial Attacks on Vision-Language-Action Models (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