Harness-Search: Guiding Long-Horizon Search through Multi-Agent Coordination

Shanyong Wang; Zhenwen Ji; Lei Jin; Yining Zhao; Yicheng Qian; Chengqiang Lu; Yi Wu; Yao Hu

cs.CLOct 4, 2026

Harness-Search: Guiding Long-Horizon Search through Multi-Agent Coordination

Authors: Shanyong Wang, Zhenwen Ji, Lei Jin, Yining Zhao, Yicheng Qian, Chengqiang Lu, Yi Wu, Yao Hu, +2 more

Organizations: Xiaohongshu Inc. · the Joint SDU-NTU Centre for Artificial Intelligence Research (C-FAIR), Shandong University · University of Illinois at Urbana-Champaign

Abstract

Long-horizon search requires agents to gather evidence across multiple steps and synthesize it into well-supported answers. The recent agent harnesses provide a natural and promising framework to support such long-running search processes. As interaction histories grow, one single agent in harnesses might get stuck and cause the policy to lose track of unresolved questions, overlook useful evidence, or terminate before sufficient support has been collected. One of promising way is to decouple three distinct responsibilities of proposing retrieval actions, updating persistent state, and deciding when to stop rather than concentrating them within a single policy. Targeted at it, we introduce Harness-Search, a multi-agent search harness to reduce the local errors propagating across subsequent exploration, evidence curation, and termination decisions. In particular, Harness-Search assigns these responsibilities to three permission-bounded authorities: a Retrieval Policy that proposes search operations, a Memory Operator that validates and commits persistent-state updates, and a Summary Auditor that accepts or rejects termination based on the sufficiency of the curated evidence. Together, these roles form a Propose-Commit-Audit loop in which actions are proposed, persistent evidence is selectively committed, and stopping decisions are subjected to an explicit sufficiency check. Across seven long-horizon search benchmarks, Harness-Search improves both retrieval and answer generation under the same policy backbone, increasing Recall by 4.60-27.92 points and Final-Answer Recall by 12.34-30.13 points over the strongest harness-based baseline on each evidence-retrieval benchmark. Moreover, trajectory-level analyses show that Harness-Search continues to accumulate useful evidence and expand evidence coverage with less redundant retrieval as the search history grows.

Figures & tables

Figure 1: Harness-Search turns open-ended search into a guided, evidence-driven process. Naive iterative search often gets stuck exploring the same direction, leading to redundant retrieval and missing evidence. Harness-Search explicitly alternates between planning, searching, curating, and auditing, and redirects the search when the auditor identifies an evidence gap. As a result, it continually expands search coverage instead of saturating early ( Bottom Right ), yielding substantial improvements across diverse search-intensive benchmarks ( Bottom Left ).

Figure 2: Overview of Harness-Search . Harness-Search addresses long-horizon search with three permission-bounded agents: a Retrieval Policy proposes actions, a Memory Operator validates and commits persistent state, and a Summary Auditor audits evidence sufficiency before termination.

Table 3

Figure 4: Behavior following a Summary Auditor rejection.

Appendix figures & tables9 assets

Supplementary material from the paper’s appendix.

Appendix

Tool signature	Effect and output
fan_out_search(queries)	Executes up to five complementary hybrid searches and returns the combined ranked document IDs and text snippets.
search(query)	Performs a single hybrid search using sparse and dense retrieval, followed by fusion and reranking; returns ranked document IDs and text snippets.
grep_corpus(pattern)	Matches an exact or regular-expression pattern against the corpus and returns matching document IDs and text excerpts.
read(doc_id)	Retrieves the full content of the specified document and adds it to the document memory.
review_docs(doc_ids)	Re-renders up to five previously retrieved documents from outer memory without issuing an additional corpus call.
redirect(doc_ids, reasoning)	Sends up to five newly retrieved candidate documents and the policy’s rationale to the intent model; returns an updated active retrieval intent.

Appendix

Table 6: Intent Curator tool signatures.

Seed	BC+	Web	Sec	LongSeal
20	55.00	59.20	35.20	86.00
42	57.81	62.19	37.97	88.98
314	61.50	65.50	41.00	92.20
Avg.	$58.10{\color[rgb]{0.5,0.5,0.5}\scriptstyle\pm 3.26}$	$62.30{\color[rgb]{0.5,0.5,0.5}\scriptstyle\pm 3.15}$	$38.06{\color[rgb]{0.5,0.5,0.5}\scriptstyle\pm 2.90}$	$89.06{\color[rgb]{0.5,0.5,0.5}\scriptstyle\pm 3.10}$
42	56.32	61.05	36.41	87.52
42	58.27	62.48	37.83	89.17

Appendix

Table 7: Repeated experiments across same and random seeds

Category	System or intervention	$R_{\mathrm{ablation}}$	$\Delta R$ (95% CI)	FA-R	TR
Full	–	57.81	—	61.51	66.86
Component	Without Memory Operator	44.74	$+13.07\ [10.71,15.43]$	49.89	59.08
	Without Summary Auditor	52.07	$+5.74\ [3.52,7.90]$	56.67	65.34
	Without Document Relevance Judge	52.98	$+4.83\ [2.78,6.86]$	55.99	64.26
	Without Direction Planner	52.58	$+5.23\ [3.52,6.94]$	54.43	63.12
Monolithic	True single agent + structured tools	54.40	$+3.41\ [1.29,5.55]$	55.93	63.34

Appendix

Table 8: Full Multi-agent decomposition study.

R_{\mathrm{ctrl}}

is the Recall of each control.

\Delta R=R_{\mathrm{full}}-R_{\mathrm{ablation}}

; positive values favor the complete system. FA-R denotes final-answer Recall, TR denotes trajectory Recall. All Recall values are percentages.

Panel of Figure 5 — Figure 5: Sensitivity to inference-time interaction.

Max Turn	R	FA-R	TR	Mean latency(s)	P50	P95	P99
10	44.74	49.12	51.90	27.03	25.13	47.78	61.23
20	54.69	57.74	61.79	39.42	37.93	63.96	88.94
30	56.55	59.36	63.81	47.95	46.68	83.89	99.88
40	57.81	61.51	66.86	52.15	52.50	90.79	114.96
50	57.77	63.11	68.59	60.34	59.76	112.19	138.69
60	56.40	60.88	66.12	67.91	68.95	129.04	166.30

Appendix

Table 9: Sensitivity to the total interaction budget and the search-before-curation budget on BrowseComp-Plus using GPT-OSS-20B. In the upper block, we vary the maximum number of total interaction turns

T_{\max}

while fixing

S_{\max}=5

. In the lower block, we vary the maximum number of consecutive search actions before a required curate operation,

S_{\max}

, while fixing

T_{\max}=40

. R, FA-R, and TR denote final Recall, final-answer Recall, and trajectory Recall, respectively. We select

T_{\max}=40

and

S_{\max}=5

for the main experiments as a quality–latency trade-off.

Figure 6: Role-specific model routing on BrowseComp-Plus. The Retrieve Policy is fixed to GPT-OSS-20B, while the Memory Operator (columns) and Summary Auditor (rows) are independently assigned GPT-OSS-20B or Qwen3 models of different sizes. Panels report (a) Recall of the final curated set, (b) final-answer Recall over gold documents, and (c) trajectory Recall over documents encountered during search. Values are percentages, with darker colors indicating higher scores.

Figure 7: Training loss figures of three agents.

Figure 8: Case Study 1. Compared to other method(Left), the improvement mainly comes from Redirect by Retrieve Policy and search_more by Summary Auditor.

Figure 9: Case Study 2. The step-by-step results show that most of the improvement comes from Redirect decisions made by Retrieval Policy.

Explore similar work

Harness-1: Reinforcement Learning for Search Agents with State-Externalizing Harnesses
Jun 1, 2026Pengcheng Jiang, Zhiyi Shi, Kelly Hong +5Search Agents Retrievers
University of Illinois at Urbana-Champaign · †UC Berkeley · ✧Chroma
LexiHorizon: Stabilizing Reinforcement Learning for Long-Horizon Deep Search
Oct 4, 2026Zhiqing Nong, Liang Wen, Chao-Hsuan LiuSearch Agents
Harness-G: A Graph-Structured Harness for Search Agents
Jul 30, 2026Yanning Hou, Haoyuan Chen, Sihang Zhou +7Search Agents Credit Assignment
National University of Defense Technology · Changsha, China

Jun 1, 2026, cs.AIJ/K move · Enter open · S save

Harness-1: Reinforcement Learning for Search Agents with State-Externalizing Harnesses

Pengcheng Jiang, Zhiyi Shi, Kelly Hong, Xueqiang Xu+4

University of Illinois at Urbana-Champaign · †UC Berkeley · ✧Chroma

Search Agents Retrievers

Open paper AskPDF arXiv

Search agents are often trained as policies over growing transcripts: the model must decide how to search while also remembering what it has seen, which evidence is useful, which constraints remain open, and which claims have actually been checked. We argue that this formulation puts too much routine state management inside the policy: reinforcement learning is forced to optimize both semantic search decisions and recoverable bookkeeping that the environment can maintain more reliably. We introduce Harness-1, a 20B search agent (retrieval subagent) trained with reinforcement learning inside a stateful search harness. The harness maintains environment-side working memory, including a candidate pool, an importance-tagged curated set, compact evidence links, verification records, compressed and deduplicated observations, and budget-aware context rendering. The policy retains the semantic decisions: what to search, which documents to keep or discard, what to verify, and when to stop. Across eight retrieval benchmarks spanning web, finance, patents, and multi-hop QA, Harness-1 achieves 0.730 average curated recall, outperforming the next strongest open search subagent by +11.4 points and remaining competitive with much larger frontier-model searchers. Its gains are especially strong on held-out transfer benchmarks, suggesting that reinforcement learning over explicit search state can produce retrieval behaviors that generalize beyond the training domains. Our code is available at https://github.com/pat-jj/harness-1.

Abstract

Explore similar work

Harness-1: Reinforcement Learning for Search Agents with State-Externalizing Harnesses

LexiHorizon: Stabilizing Reinforcement Learning for Long-Horizon Deep Search

Harness-G: A Graph-Structured Harness for Search Agents