Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

SimpleSearch-VL: A Simple Recipe for Multimodal Agentic Deep Search

About

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.

Ming Dai, Zhihong Lu, Jinjie Gu, Jiedong Zhuang, Yefeng Liu, Wankou Yang, Jian Wang, Chunhua Shen• 2026

Related benchmarks

TaskDatasetResultRank
Visual Question AnsweringLiveVQA--
151
Visual Question AnsweringFVQA
Accuracy79
57
Visual Question AnsweringSimpleVQA
Score79.6
38
Multimodal SearchBrowseComp-VL
Score55.9
35
Multimodal Agentic SearchMMSearch
Score83.6
25
Multimodal Agentic SearchMMSearch+
Score34.4
19
Showing 6 of 6 rows

Other info

Follow for update