{"slug": "think-big-search-small-where-capacity-matters-in-hierarchical-search-agents", "title": "Think Big, Search Small: Where Capacity Matters in Hierarchical Search Agents?", "summary": "Researchers at an undisclosed institution found that in hierarchical search agents using large language models, scaling the delegation role improves exact match by ~11 points while scaling execution only improves by ~2.6 points, identifying task decomposition as the capability bottleneck. A 1.7B-parameter executor trained via trajectory distillation matched frontier accuracy with 37% fewer tokens, suggesting capacity should be concentrated at delegation.", "body_md": "arXiv:2607.07548v1 Announce Type: new\nAbstract: Large language model based search agents increasingly adopt multi-agent architectures in which a main agent decomposes a complex question into sub-queries and dispatches them to parallel sub-agents. However, existing systems instantiate all roles from a single model of identical scale, leaving open how model capacity should be distributed across roles. We factorize hierarchical search into three roles: a delegation role responsible for task decomposition, an execution role responsible for retrieval and evidence extraction, and an answer generation role held fixed as a confound control. We then conduct controlled capacity sweeps along the delegation and execution axes on five multi-hop QA benchmarks. The experiments yield three findings. First, role factorization consistently outperforms a single-agent baseline, improving exact match from 4.5 to 8.6 points across six model scales. Second, capacity sensitivity is asymmetric: scaling the delegation backbone improves EM by ~11 points, whereas scaling the execution sub-agent moves EM by only ~2.6 points, identifying decomposition as the capability bottleneck. Third, a 1.7B-parameter executor trained via quality-filtered trajectory distillation matches a frontier sub-agent in accuracy while consuming 37% fewer sub-agent tokens, advancing the Pareto frontier. These results suggest a concrete recipe for building hierarchical search agents: concentrate capacity at delegation and downsize execution without sacrificing accuracy. Our code is available at https://github.com/QinnanCai0115/role-factorized-search.", "url": "https://wpnews.pro/news/think-big-search-small-where-capacity-matters-in-hierarchical-search-agents", "canonical_source": "https://arxiv.org/abs/2607.07548", "published_at": "2026-07-09 04:00:00+00:00", "updated_at": "2026-07-09 04:16:22.271738+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-agents", "ai-research", "ai-infrastructure"], "entities": ["arXiv"], "alternates": {"html": "https://wpnews.pro/news/think-big-search-small-where-capacity-matters-in-hierarchical-search-agents", "markdown": "https://wpnews.pro/news/think-big-search-small-where-capacity-matters-in-hierarchical-search-agents.md", "text": "https://wpnews.pro/news/think-big-search-small-where-capacity-matters-in-hierarchical-search-agents.txt", "jsonld": "https://wpnews.pro/news/think-big-search-small-where-capacity-matters-in-hierarchical-search-agents.jsonld"}}