5arXiv cs.CL (Computation and Language)·
WebSwarm: Recursive multi-agent framework for deep-and-wide web search
WebSwarm is a new multi-agent orchestration framework for LLM-based web search that uses progressive recursive delegation to handle both depth and breadth simultaneously. Each agent node couples a local objective with a search mode and can either solve its task or delegate to child nodes, passing evidence upward for parent-level expansion and aggregation. The system outperforms single-agent and multi-agent baselines on BrowseComp-Plus, WideSearch, DeepWideSearch, and GISA benchmarks. The work addresses known limitations of ReAct-style single-agent search and flat parallel multi-agent approaches.
Related guides (2)
Related events (8)
Open-source DeepResearch – Freeing our search agents
Hugging Face published a blog post introducing Open Deep Research, an open-source replication of agentic deep research capabilities (similar to OpenAI's Deep Research). The project aims to build open-weight search agents capable of multi-step web research and synthesis. The post details the architecture, tooling, and early benchmark results of the system.
Copewell: Multi-agent swarm architecture for equitable mental health support
Researchers present Copewell, a multi-agent swarm system designed to expand mental wellness access, particularly in low- and middle-income countries. The architecture combines multi-source emotional assessment (self-reported, physiological, contextual), valence-arousal emotion routing via Russell's Circumplex Model, and dual-mode intervention delivery. The system incorporates a dedicated Ethics Supervisor agent and privacy-first design, with early practitioner engagement informing the architecture but no large-scale empirical evaluation yet reported.
Role-factorized hierarchical search agents: delegation capacity matters more than execution
A new arXiv paper investigates how model capacity should be distributed across roles in multi-agent search systems, factorizing hierarchical search into delegation, execution, and answer generation roles. Controlled sweeps across five multi-hop QA benchmarks find that scaling the delegation backbone improves exact match by ~11 points while scaling execution sub-agents yields only ~2.6 points, identifying task decomposition as the primary bottleneck. A 1.7B-parameter executor trained via trajectory distillation matches frontier sub-agent accuracy while using 37% fewer tokens, advancing the efficiency Pareto frontier. The results offer a concrete design recipe: concentrate capacity at delegation and downsize execution.
SwarmHarness: Decentralized Skill-Based Task Routing Protocol for AI Agent Networks
SwarmHarness is a proposed decentralized protocol enabling AI compute sharing and task routing across heterogeneous nodes (workstations, inference servers, edge devices) without a central coordinator. It combines a DHT-based registry for peer discovery, a utility-function router dispatching tasks by capability/load/latency/trust, and a Shapley-value-based credit mechanism to align incentives among participating nodes. The system is designed as a foundational primitive for autonomous multi-agent networks where agents can hire compute, route subtasks, and settle credits without human intermediation. It positions itself against existing approaches like Golem, BrokerChain, BOINC, and Petals by integrating decentralization with a native incentive layer.
Comparative Study: Semantic Metadata vs. Unstructured Web Retrieval for Agentic Data Discovery
This paper evaluates whether LLM-based agents still need structured semantic metadata (e.g., schema.org) for data retrieval, comparing a Baseline Agent searching open-web documents against a Semantic Agent leveraging 90 million schema.org-annotated datasets. Using an LLM-as-a-judge pipeline aligned to FAIR principles, the Semantic Agent achieves 65.7% higher overall precision in retrieving FAIR-compliant datasets, while the Baseline Agent answers 40% more questions but frequently returns prose-heavy or portal landing pages instead of actionable data. The study concludes that structured semantic ecosystems remain essential for reliable, execution-oriented agentic workflows despite LLMs' broad unstructured retrieval capabilities.
TradingAgents: Multi-Agent LLM Financial Trading Framework
TradingAgents is an open-source Python framework by TauricResearch that applies multi-agent LLM architectures to financial trading tasks. The repository has accumulated 81,650 GitHub stars with 284 added today, indicating strong community traction. It represents a concrete deployment pattern for agentic AI systems in quantitative finance.
SearchGEO framework measures LLM search agent vulnerability to web content manipulation
Researchers introduce SearchGEO, a controlled evaluation framework for measuring endorsement corruption in LLM-based web-search agents, combining a manipulation pipeline, five-mode attack taxonomy, and multiple output metrics. Evaluating 13 LLM backends on 308 cases each, they find attack success rates ranging from 0.0% on Claude-Sonnet-4.6 to 31.4% on Gemini-3-Flash, with model-family-specific vulnerability patterns. An auxiliary probe escalating endorsement to install commands reveals a behavioral split: Claude over-rejects while GPT over-trusts. The findings argue for treating adversarial search content robustness as a first-class safety evaluation dimension for deployed agents.
SWE-Explore: New benchmark isolates repository exploration capability in coding agents
SWE-Explore is a new benchmark targeting repository exploration as a distinct, fine-grained capability of coding agents, separate from end-to-end task resolution. It covers 848 issues across 10 programming languages and 203 open-source repositories, with line-level ground truth derived from successful agent trajectories. Evaluation across retrieval methods, coding agents, and specialized localizers finds that agentic explorers outperform classical retrieval, and that line-level coverage and efficient ranking remain the key differentiators at the frontier. The benchmark addresses a gap in SWE-bench-style evaluations that treat task resolution as a binary outcome.