Yunque DeepResearch Technical Report

TL;DR

Yunque DeepResearch introduces a hierarchical, modular framework for autonomous agents that enhances robustness, manages context effectively, and improves performance on complex deep research benchmarks.

Contribution

The paper presents a novel hierarchical, modular architecture with a multi-agent system, dynamic context management, and proactive supervision to advance autonomous deep research capabilities.

Findings

Achieves state-of-the-art results on multiple deep research benchmarks.

Demonstrates robustness and modularity in complex open-ended tasks.

Provides open-source implementation for community use.

Abstract

Deep research has emerged as a transformative capability for autonomous agents, empowering Large Language Models to navigate complex, open-ended tasks. However, realizing its full potential is hindered by critical limitations, including escalating contextual noise in long-horizon tasks, fragility leading to cascading errors, and a lack of modular extensibility. To address these challenges, we introduce Yunque DeepResearch, a hierarchical, modular, and robust framework. The architecture is characterized by three key components: (1) a centralized Multi-Agent Orchestration System that routes subtasks to an Atomic Capability Pool of tools and specialized sub-agents; (2) a Dynamic Context Management mechanism that structures completed sub-goals into semantic summaries to mitigate information overload; and (3) a proactive Supervisor Module that ensures resilience through active anomaly detection and context pruning. Yunque DeepResearch achieves state-of-the-art performance across a range of agentic deep research benchmarks, including GAIA, BrowseComp, BrowseComp-ZH, and Humanity's Last Exam. We open-source the framework, reproducible implementations, and application cases to empower the community.