搜索超越可教範疇：演化代理視覺生成中的知識邊界

摘要

视觉生成器在渲染方面表现卓越，但对于自身不了解的内容，它们会坚定地虚构。用户请求是无边界的、持续演进的，且呈现深度的长尾分布：新角色、流行实体、截止日期后的事件等等。这种世界知识的瓶颈是结构性的：生成器基于固定的语料库训练，而视觉世界是开放式的。我们构建了 SearchGen-20K 和 SearchGen-Bench，包含 20,839 个提示词，覆盖十二个失败类别和二十二个领域，并配有一个预执行的 SearchGen-Corpus-1M 多模态语料库，以支持离线、可复现的研究。在 SearchGen-Bench 上，前沿的开放生成器得分仅为 21 到 28 分（满分 100），与现有基准相比出现了 40 分的断崖式下降，这种下降在现有基准测试中完全不可见。自然的解决方案是使用搜索工具，实现智能体式的视觉生成。然而，我们观察到简单的搜索方式会失败：它不加区分地检索，将噪声注入到生成器本身已经能处理的提示词中。我们将根本原因归结为生成器特有的、不断演变的知识边界：即生成器通过训练可以内化的知识与必须保留在外部上下文中的知识之间的分界线。尽管这一边界难以预先确定，但我们证明它可以通过“先教学、后搜索”的协同训练框架来发现。即使这个协同训练方案的最简版本也能产生单调的改进，为视觉生成中的递归式自我改进奠定基础，从而满足需要世界知识支撑的请求。我们发布了完整的数据集、协同训练语料库和搜索语料库，作为工具增强、世界知识驱动的视觉生成的可复现测试工具集。

English

Visual generators excel at rendering, but they confidently fabricate what they do not know. User requests are unbounded, evolving, and deeply long-tailed: new characters, trending entities, post-cutoff events, and more. This world-knowledge bottleneck is structural: generators are trained on fixed corpora, but the visual world is open-ended. We construct SearchGen-20K and SearchGen-Bench, with 20,839 prompts spanning twelve failure categories and twenty-two domains, paired with a pre-executed multimodal SearchGen-Corpus-1M to support offline, reproducible research. On SearchGen-Bench, frontier open generators score only 21 to 28 out of 100, a 40-point collapse invisible to existing benchmarks. The natural remedy is to employ search tools, enabling agentic visual generation. However, we find that naive search fails: it retrieves indiscriminately, injecting noise into prompts the generator already handles. We trace the root cause to a generator-specific, evolving knowledge boundary: the divide between what a generator can internalize through training and what must remain in external context. Although this boundary is hard to specify in advance, we show that it is discoverable through a teach-then-search co-training framework. Even a minimal version of this co-training recipe produces monotonic improvement, laying the foundation for recursive self-improvement in visual generation that can meet world-knowledge-grounded requests. We release the full dataset, co-training corpus, and search corpus as a replayable harness for tool-augmented, world-knowledge-grounded visual generation.