VSearcher: Long-Horizon Multimodal Search Agent via Reinforcement Learning

TL;DR

VSearcher is a reinforcement learning-based multimodal search agent capable of long-horizon, multi-turn interactions with web environments, significantly improving multimodal web search performance over existing models.

Contribution

The paper introduces VSearcher, a novel multimodal search agent with a new data synthesis pipeline and a specialized benchmark for evaluating multimodal search capabilities.

Findings

VSearcher outperforms recent multimodal search agents.

VSearcher surpasses some proprietary models on web search tasks.

The proposed methods improve multi-turn, multimodal web interaction.

Abstract

Large models are increasingly becoming autonomous agents that interact with real-world environments and use external tools to augment their static capabilities. However, most recent progress has focused on text-only large language models, which are limited to a single modality and therefore have narrower application scenarios. On the other hand, multimodal large models, while offering stronger perceptual capabilities, remain limited to static knowledge and lack the ability to access and leverage up-to-date web information. In this paper, we propose VSearcher, turning static multimodal model into multimodal search agent capable of long-horizon, multi-turn tool use in real-world web environments, including text search, image search, and web browsing, via reinforcement learning. Specifically, we introduce Iterative Injection Data Synthesis pipeline to generate large-scale, complex multimodal QA questions, which are further filtered with comprehensive metrics to ensure high quality and sufficient difficulty. We then adopt an SFT-then-RL training pipeline to turn base multimodal models to agent capable of multi-turn tool calling in real-world web environments. Besides, we propose a multimodal search benchmark MM-SearchExam dedicated to evaluating search capabilities of multimodal search agents, which proves highly challenging for recent proprietary models. Extensive evaluations across multiple multimodal search benchmarks reveal effectiveness of our method. VSearcher achieves superior performance compared to recent multimodal search agents and even surpasses several proprietary models on multimodal web search tasks.