---
title: "WebSwarm: Recursive Multi-Agent Orchestration for Deep-and-Wide Web Search"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2607.08662v1"
permalink: "https://feed.astralor.com/items/itm_ce8fdbfde4c93860659929e8"
published: "2026-07-09T16:28:49.000Z"
score: 78
channel: "ai"
tags: ["WebSwarm","Multi-agent Systems","Large Language Models","Web Search","Recursive Delegation"]
---

# WebSwarm: Recursive Multi-Agent Orchestration for Deep-and-Wide Web Search

## AI commentary

> This paper is worth attention as it tackles key bottlenecks in existing multi-agent search systems regarding recursive depth and dynamic collaboration. For researchers and developers working on AI agents, knowledge retrieval, and complex reasoning, its ideas of process-level experience reuse and structured delegation are highly insightful.

## At a glance

WebSwarm is a progressive recursive delegation framework that dynamically instantiates search nodes coupling local objectives with search modes, enabling task decomposition, recursive expansion, and agent collaboration. Experiments on multiple deep-and-wide search benchmarks show WebSwarm consistently outperforms single-agent and existing multi-agent baselines, effectively addressing the trade-off between depth and coverage.

## Key points

- Proposes WebSwarm framework that enhances both search depth and coverage via recursive delegation and collaboration
- Dynamically instantiates search nodes, each can solve locally or delegate to child nodes
- First probes web information organization to guide node expansion and reuse process-level experience
- Achieves state-of-the-art performance on BrowseComp-Plus, WideSearch, and other benchmarks

## Full text

Large language model (LLM)-based web search agents are transforming information seeking from simple factoid question answering into complex, deep-and-wide search and research-oriented tasks. A single ReAct-style agent is constrained by one long trajectory and limited context, making it difficult to handle depth and coverage simultaneously. Existing multi-agent systems improve search coverage through parallel execution and aggregation, but still exhibit clear limitations in recursive depth, collaboration adaptability, and evidence-grounded expansion. We propose WebSwarm, a progressive recursive delegation framework that jointly constructs task decomposition, recursive expansion, and agent collaboration during inference. WebSwarm dynamically instantiates agentic search nodes, each coupling a local objective with a search mode that specifies how the node should organize search and collaboration. Each node can either solve its objective itself or further delegate child nodes; after solving, it returns evidence and results upward, enabling parent nodes to further expand, revise, or aggregate the search process. To guide this process, WebSwarm first probes how task-relevant information is organized on the web to ground subsequent node expansion, and reuses process-level experience across homogeneous sibling nodes. Experiments on BrowseComp-Plus, WideSearch, DeepWideSearch, and GISA show that WebSwarm consistently outperforms single-agent and multi-agent baselines on deep, wide, and interleaved deep-and-wide tasks. Further analyses of ablation, task difficulty, web tool efficiency, and model generalization explain WebSwarm's effectiveness and provide insights for multi-agent search systems.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_ce8fdbfde4c93860659929e8