用轻量框架提升多模态智能体搜索能力，效果媲美顶级模型。
SimpleSearch-VL: A Simple Recipe for Multimodal Agentic Deep Search
- 通过自适应采样优化搜索效率，用冗余样本降低长尾延迟。
- 引入证据验证机制，显式评估图文信息与上下文相关性。
- 无需外部依赖，轻量工具接口实现网页自动摘要，适合落地部署。
我们提出 SimpleSearch-VL，一个高效、可靠且实用的多模态智能体搜索框架。其核心思想是优化智能体自身的搜索与验证过程，而非扩展数据、工具或辅助模型组件。为提升效率，采用分治自适应回溯（FAR）策略，在保持高信息量训练组的同时，利用冗余样本缓解长尾延迟并暴露难例。为增强可靠性，实施证据验证推理，显式使用思维链验证检索到的图文线索与原始上下文的相关性。为提高实用性，框架保持轻量级工具接口，并在智能体内完成网页自摘要，无需额外外部依赖。仅需5,000条监督工具交错轨迹和2,000条强化学习数据，SimpleSearch-VL在Qwen3-VL的8B和30B-A3B变体上分别提升基线15.8和16.0平均分。SimpleSearch-VL-30B-A3B模型性能已达到与Gemini-3-Pro相当水平。
原文摘要 · Abstract (English)
We present SimpleSearch-VL, an efficient, reliable, and practical framework for multimodal agentic search. Its core idea is to improve the agent's own search-and-verification process rather than scaling data, tools, or auxiliary model components. For efficiency, Factorized Adaptive Rollout (FAR) improves sampling efficiency by forming more informative training groups while using redundant samples to mitigate long-tail latency and expose hard samples. For reliability, SimpleSearch-VL performs evidence-verified reasoning, explicitly using chain-of-thought verification to assess the relevance of retrieved visual and textual cues to the original context. For practicality, SimpleSearch-VL keeps a lightweight tool interface and performs webpage self-summary within the agent, requiring no additional external dependencies. With only 5K supervised tool-interleaved trajectories and 2K RL data, SimpleSearch-VL improves Qwen3-VL agentic baselines by 15.8 and 16.0 average points for the 8B and 30B-A3B variants, respectively. The SimpleSearch-VL-30B-A3B model further achieves performance competitive with agentic Gemini-3-Pro.
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。