让统一多模态模型自动生成对抗样本，提升理解与生成的一致性。
UniGame: Turning a Unified Multimodal Model Into Its Own Adversary
- 在共享令牌层加入轻量扰动器，让生成分支主动挑战理解分支。
- 理解能力提升3.6%，生成性能提升0.02，在分布外和对抗攻击下鲁棒性增6.2%。
- 框架不依赖架构、参数增加不足1%，适合各类多模态模型增强。
统一多模态模型（UMMs）在单一架构下展现出出色的理解和生成能力。然而，其仍存在根本性矛盾：理解偏好紧凑嵌入，生成则依赖高重建信息表示。这种结构权衡导致决策边界错位、跨模态一致性下降，并在分布外和对抗性扰动下更易失效。本文提出UniGame，一种自对抗后训练框架，直接针对这一不一致问题。通过在共享令牌接口施加轻量扰动器，使生成分支主动寻找并挑战理解分支的脆弱点，从而将模型自身转化为其对手。实验表明，UniGame显著提升了模型一致性（+4.6%），理解能力（+3.6%）、生成性能（+0.02，于GenEval），以及在分布外和对抗鲁棒性上分别提升4.8%（NaturalBench）和6.2%（AdVQA）。该框架具备架构无关性，参数增量低于1%，且可与现有后训练方法互补。结果表明，对抗自博弈是未来多模态基础模型实现更高一致性、稳定性与统一能力的有效通用原则。官方代码见：https://github.com/AIFrontierLab/TorchUMM
原文摘要 · Abstract (English)
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)on GenEval, 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/TorchUMM
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。