RealUnify：统一模型真的从统一中获益吗？一个全面的基准
RealUnify: Do Unified Models Truly Benefit from Unification? A Comprehensive Benchmark
对做统一多模态模型（理解+生成）的研究者，这个基准提供了可复用的双向协同评测协议：把任务拆成理解阶段与生成阶段分别打分，能定位瓶颈是核心能力不足还是能力整合失败，可直接用来诊断自家统一模型的训练策略与归纳偏置设计。
原文依据 a benchmark specifically designed to evaluate bidirectional capability synergy
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论将视觉理解与生成整合进统一多模态模型，是迈向通用人工智能的重要一步。然而，现有基准仍未回答一个根本性问题：这种架构层面的统一，是否真的能让各项组成能力之间产生协同互动？现有评测范式主要对理解与生成分别孤立评估，不足以判定统一模型能否利用其理解能力来增强生成，或借助生成式模拟来促进更深层的理解。为填补这一关键空白，我们提出 RealUnify，一个专门用于评估双向能力协同的基准。RealUnify 包含 1,000 个人工精细标注的实例，覆盖 10 个类别与 32 个子任务。它围绕两条核心轴线构建：1）理解增强生成，要求模型运用推理（如常识、逻辑）来指导图像生成；2）生成增强理解，要求模型进行心智模拟或重建（例如对经过变换或顺序打乱的视觉输入），以完成推理任务。我们的关键贡献是双重评测协议，它将直接的端到端评估与诊断式的分步评估相结合，后者把任务分解为相互独立的理解阶段与生成阶段。该协议使我们能够精确辨别性能瓶颈究竟源于核心能力的欠缺，还是源于能力整合的失败。通过对 12 个领先的统一模型与 6 个专用基线进行大规模评估，我们发现当前统一模型仍难以实现有效协同，这表明仅靠架构统一并不充分。这些结果凸显出，需要新的训练策略与归纳偏置，才能充分释放统一建模的潜力。
英文原文摘要 1909 字符
The integration of visual understanding and generation into unified multimodal models represents a significant stride toward general-purpose AI. However, a fundamental question remains unanswered by existing benchmarks: does this architectural unification actually enable synergetic interaction between the constituent capabilities? Existing evaluation paradigms, which primarily assess understanding and generation in isolation, are insufficient for determining whether a unified model can leverage its understanding to enhance its generation, or use generative simulation to facilitate deeper comprehension. To address this critical gap, we introduce RealUnify, a benchmark specifically designed to evaluate bidirectional capability synergy. RealUnify comprises 1,000 meticulously human-annotated instances spanning 10 categories and 32 subtasks. It is structured around two core axes: 1) Understanding Enhances Generation, which requires reasoning (e.g., commonsense, logic) to guide image generation, and 2) Generation Enhances Understanding, which necessitates mental simulation or reconstruction (e.g., of transformed or disordered visual inputs) to solve reasoning tasks. A key contribution is our dual-evaluation protocol, which combines direct end-to-end assessment with a diagnostic stepwise evaluation that decomposes tasks into distinct understanding and generation phases. This protocol allows us to precisely discern whether performance bottlenecks stem from deficiencies in core abilities or from a failure to integrate them. Through large-scale evaluations of 12 leading unified models and 6 specialized baselines, we find that current unified models still struggle to achieve effective synergy, indicating that architectural unification alone is insufficient. These results highlight the need for new training strategies and inductive biases to fully unlock the potential of unified modeling.