← Back ◬ AI & Machine Learning Jun 17, 2026

Beyond Parallel Sampling: Diverse Query Initialization for Agentic Search

arXiv AI Archived Jun 17, 2026 ✓ Full text saved

arXiv:2606.17209v1 Announce Type: new Abstract: Test-time scaling for agentic search typically increases depth (i.e., more turns and tokens per trajectory) or breadth (i.e., more parallel rollouts). Here we focus on breadth scaling, showing that standard parallel sampling yields diminishing returns, tracing this to query redundancy at the first turn. When models issue similar first queries across rollouts, the threads retrieve overlapping evidence, and subsequent turns are conditioned on this sh

Full text archived locally

✦ AI Summary · Claude Sonnet

Computer Science > Artificial Intelligence [Submitted on 15 Jun 2026] Beyond Parallel Sampling: Diverse Query Initialization for Agentic Search Sidhaarth Murali, João Coelho, Jingjie Ning, João Magalhães, Bruno Martins, Chenyan Xiong Test-time scaling for agentic search typically increases depth (i.e., more turns and tokens per trajectory) or breadth (i.e., more parallel rollouts). Here we focus on breadth scaling, showing that standard parallel sampling yields diminishing returns, tracing this to query redundancy at the first turn. When models issue similar first queries across rollouts, the threads retrieve overlapping evidence, and subsequent turns are conditioned on this shared retrieval. We address this limitation with DivInit, a training-free intervention at the first turn. Rather than sampling k independent first queries, DivInit draws n candidates from a single call, picks k < n diverse seeds, and runs them as parallel trajectories. Across five open-weight models and eight benchmarks, DivInit consistently improves over standard parallel sampling, with average gains of five to seven points on multi-hop QA at matched compute. Code available at this https URL Comments: 15 pages, 8 figures; under review at EMNLP 2026 Subjects: Artificial Intelligence (cs.AI); Information Retrieval (cs.IR) Cite as: arXiv:2606.17209 [cs.AI] (or arXiv:2606.17209v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2606.17209 Focus to learn more Submission history From: Sidhaarth Sredharan [view email] [v1] Mon, 15 Jun 2026 18:48:11 UTC (652 KB) Access Paper: HTML (experimental) view license Current browse context: cs.AI < prev | next > new | recent | 2026-06 Change to browse by: cs cs.IR References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

💬 Team Notes