QuarkMedSearch: A Long-Horizon Deep Search Agent for Exploring Medical Intelligence

2026-04-14

AI Summary

This work addresses the limited performance of foundation models in Chinese medical vertical domains for deep search tasks by proposing a full-stack optimization framework. The approach integrates large-scale medical knowledge graphs with online exploration, introduces a novel multi-hop medical data synthesis strategy, and enhances the agent’s planning, tool-use, and reflection capabilities through a two-stage fine-tuning process combining supervised fine-tuning (SFT) and reinforcement learning (RL). The study also establishes QuarkMedSearch Benchmark—the first expert-validated Chinese medical deep search evaluation benchmark—on which the proposed method achieves state-of-the-art performance among open-source models of comparable scale while maintaining strong competitiveness on general-purpose tasks.

Technology Category

• Search and Optimization: Learning to Search
• Multiagent Systems: Other Foundations of Multi Agent Systems
• Machine Learning: Deep Neural Architectures and Foundation Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

As agentic foundation models continue to evolve, how to further improve their performance in vertical domains has become an important challenge. To this end, building upon Tongyi DeepResearch, a powerful agentic foundation model, we focus on the Chinese medical deep search scenario and propose QuarkMedSearch, systematically exploring a full-pipeline approach spanning medical multi-hop data construction, training strategies, and evaluation benchmarks to further push and assess its performance upper bound in vertical domains. Specifically, for data synthesis, to address the scarcity of deep search training data in the medical domain, we combine a large-scale medical knowledge graph with real-time online exploration to construct long-horizon medical deep search training data; for post-training, we adopt a two-stage SFT and RL training strategy that progressively enhances the model's planning, tool invocation, and reflection capabilities required for deep search, while maintaining search efficiency; for evaluation, we collaborate with medical experts to construct the QuarkMedSearch Benchmark through rigorous manual verification. Experimental results demonstrate that QuarkMedSearch achieves state-of-the-art performance among open-source models of comparable scale on the QuarkMedSearch Benchmark, while also maintaining strong competitiveness on general benchmarks.

Problem

• agentic foundation models
• vertical domains
• medical deep search
• long-horizon reasoning
• performance upper bound

Innovation

• long-horizon deep search
• medical knowledge graph
• two-stage SFT and RL
• multi-hop data synthesis
• domain-specific benchmark