MM-DeepResearch: A Simple and Effective Multimodal Agentic Search Baseline

We aim to develop a multimodal research agent capable of explicit reasoning and planning, multi-tool invocation, and cross-modal information synthesis, enabling it to conduct deep research tasks. However, we observe three main challenges in developing such agents: (1) scarcity of search-intensive multimodal QA data, (2) lack of effective search trajectories, and (3) prohibitive cost of training with online search APIs. To tackle them, we first propose Hyper-Search, a hypergraph-based QA generation method that models and connects visual and textual nodes within and across modalities, enabling to generate search-intensive multimodal QA pairs that require invoking various search tools to solve. Second, we introduce DR-TTS, which first decomposes search-involved tasks into several categories according to search tool types, and respectively optimize specialized search tool experts for each tool. It then recomposes tool experts to jointly explore search trajectories via tree search, producing trajectories that successfully solve complex tasks using various search tools. Third, we build an offline search engine supporting multiple search tools, enabling agentic reinforcement learning without using costly online search APIs. With the three designs, we develop MM-DeepResearch, a powerful multimodal deep research agent, and extensive results shows its superiority across benchmarks.

We develop MM-DeepResearch, a multimodal AI research assistant that can look through text and images, decide which tools to use, and combine information from different sources to answer complex questions. Building such a system is difficult because there is little training data for realistic search-heavy tasks, good examples of multi-step search behavior are hard to obtain, and training with real online search services is expensive. To address this, we create new methods to generate challenging question-answer data, train specialized search skills for different tools, and simulate search in an offline environment. Together, these advances allow our system to perform deep research more effectively and at lower cost, and experiments show that it outperforms existing approaches on a range of benchmarks.