SearchOS-V1: Towards Robust Open-Domain Information-Seeking Agent Collaboration

Hugging Face Daily Papers Papers
multi-agent information-seeking tool-integrated web-search collaboration open-domain robust

Summary

Introduces SearchOS, a multi-agent framework for robust open-domain information-seeking that externalizes search progress into explicit states via a novel Search-Oriented Context Management (SOCM) system, achieving state-of-the-art results on WideSearch and GISA benchmarks.

Recent advances in Tool-Integrated Large Language Models have made web search a core capability of information-seeking agents. However, as interaction histories grow, agents increasingly struggle to track task progress. When search attempts fail to yield useful evidence, current single- and multi-agent systems can become trapped in repetitive loops, wasting search budgets and ultimately compromising the quality and completeness of the final output. We introduce SearchOS, a system-level multi-agent framework that turns fragile, implicit search progress into explicit, persistent, and shared state. First, we formulate open-domain information seeking as relational schema completion with grounded citations, where agents discover entities, populate attributes across linked tables, and anchor each value to source evidence. Then we design Search-Oriented Context Management (SOCM), which externalizes the evolving state into Frontier Task, an Evidence Graph, a Coverage Map, and Failure Memory. Built on SOCM, SearchOS applies a pipeline-parallel scheduling mechanism that overlaps the execution of sub-agents and continuously refills freed slots with tasks targeting unresolved coverage gaps to improve utilization and throughput. To schedule and control the execution of search agents, SearchOS introduces a Search Tool Middleware Harness that intercepts model and tool interactions to record grounded evidence and react to stalls or budget exhaustion, and provides a reusable hierarchical skill system comprising strategy and access skills to augment the agents' search process and avoid repeating failed search patterns across runs. On WideSearch and GISA, SearchOS leads all metrics among the evaluated single- and multi-agent baselines, paving the way toward robust information-seeking collaboration.
Original Article
View Cached Full Text

Cached at: 07/20/26, 09:43 AM

Paper page - SearchOS-V1: Towards Robust Open-Domain Information-Seeking Agent Collaboration

Source: https://huggingface.co/papers/2607.15257 Authors:

,

,

,

,

,

,

,

,

,

,

,

,

Abstract

RecentadvancesinTool-IntegratedLargeLanguageModelshavemadewebsearchacorecapabilityofinformation-seekingagents.However,asinteractionhistoriesgrow,agentsincreasinglystruggletotracktaskprogress.Whensearchattemptsfailtoyieldusefulevidence,currentsingle-andmulti-agentsystemscanbecometrappedinrepetitiveloops,wastingsearchbudgetsandultimatelycompromisingthequalityandcompletenessofthefinaloutput.WeintroduceSearchOS,asystem-levelmulti-agentframeworkthatturnsfragile,implicitsearchprogressintoexplicit,persistent,andsharedstate.First,weformulateopen-domaininformationseekingasrelationalschemacompletionwithgroundedcitations,whereagentsdiscoverentities,populateattributesacrosslinkedtables,andanchoreachvaluetosourceevidence.ThenwedesignSearch-OrientedContextManagement(SOCM),whichexternalizestheevolvingstateintoFrontierTask,anEvidenceGraph,aCoverageMap,andFailureMemory.BuiltonSOCM,SearchOSappliesapipeline-parallelschedulingmechanismthatoverlapstheexecutionofsub-agentsandcontinuouslyrefillsfreedslotswithtaskstargetingunresolvedcoveragegapstoimproveutilizationandthroughput.Toscheduleandcontroltheexecutionofsearchagents,SearchOSintroducesaSearchToolMiddlewareHarnessthatinterceptsmodelandtoolinteractionstorecordgroundedevidenceandreacttostallsorbudgetexhaustion,andprovidesareusablehierarchicalskillsystemcomprisingstrategyandaccessskillstoaugmenttheagents’searchprocessandavoidrepeatingfailedsearchpatternsacrossruns.OnWideSearchandGISA,SearchOSleadsallmetricsamongtheevaluatedsingle-andmulti-agentbaselines,pavingthewaytowardrobustinformation-seekingcollaboration.

View arXiv pageView PDFGitHub65Add to collection

Get this paper in your agent:

hf papers read 2607\.15257

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2607.15257 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2607.15257 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2607.15257 in a Space README.md to link it from this page.

Collections including this paper2

Similar Articles

SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery

arXiv cs.AI

SwarmResearch introduces an orchestrator-subagent harness where a Shepherd Agent steers a population of Search Agents to explore diverse solutions for open-ended optimization problems, achieving better or comparable results to state-of-the-art methods on 13/15 tasks.