Relevance no longer only decides what enters the context. It guides where interaction begins, which documents are searched first, and which local excerpts remain visible.</p>\n<p><a href=\"https://cdn-uploads.huggingface.co/production/uploads/6891bbff78946201296b4592/ebReVrJEU5iEyQYfebh2w.png\" rel=\"nofollow\"><img src=\"https://cdn-uploads.huggingface.co/production/uploads/6891bbff78946201296b4592/ebReVrJEU5iEyQYfebh2w.png\" alt=\"cost\"></a></p>\n<p>🌟 Accuracy/nDCG@10 versus interaction cost (average tool calls) on BrowseComp-Plus and BRIGHT. By turning relevance into an execution prior over rg exploration, RARG advances the accuracy--efficiency frontier over retrieval-based and direct-interaction agents.🌟</p>\n","updatedAt":"2026-07-29T05:11:00.679Z","author":{"_id":"6891bbff78946201296b4592","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6891bbff78946201296b4592/ECmWBrlfeonPg0HzmQ_sW.png","fullname":"Yuqing Li","name":"MindscapeRAG","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":12,"isUserFollowing":false}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.8848718404769897},"editors":["MindscapeRAG"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/6891bbff78946201296b4592/ECmWBrlfeonPg0HzmQ_sW.png"],"reactions":[],"isReport":false}},{"id":"6a698b16aa5e7f40e9fe2676","author":{"_id":"6891bbff78946201296b4592","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6891bbff78946201296b4592/ECmWBrlfeonPg0HzmQ_sW.png","fullname":"Yuqing Li","name":"MindscapeRAG","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":12,"isUserFollowing":false},"createdAt":"2026-07-29T05:09:42.000Z","type":"comment","data":{"edited":true,"hidden":true,"hiddenBy":"","hiddenReason":"Spam","latest":{"raw":"This comment has been hidden","html":"This comment has been hidden","updatedAt":"2026-07-29T05:11:16.081Z","author":{"_id":"6891bbff78946201296b4592","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6891bbff78946201296b4592/ECmWBrlfeonPg0HzmQ_sW.png","fullname":"Yuqing Li","name":"MindscapeRAG","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":12,"isUserFollowing":false}},"numEdits":1,"editors":[],"editorAvatarUrls":[],"reactions":[]}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.24223","authors":[{"_id":"6a69694a9d3a1231d492b859","name":"Jiangnan Li","hidden":false},{"_id":"6a69694a9d3a1231d492b85a","name":"Yuqing Li","hidden":false},{"_id":"6a69694a9d3a1231d492b85b","name":"Mo Yu","hidden":false},{"_id":"6a69694a9d3a1231d492b85c","name":"Jinchao Zhang","hidden":false},{"_id":"6a69694a9d3a1231d492b85d","name":"Jie Zhou","hidden":false}],"publishedAt":"2026-07-27T00:00:00.000Z","submittedOnDailyAt":"2026-07-29T00:00:00.000Z","title":"A New Role for Relevance: Guiding Corpus Interaction in Agentic Search","submittedOnDailyBy":{"_id":"6891bbff78946201296b4592","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6891bbff78946201296b4592/ECmWBrlfeonPg0HzmQ_sW.png","isPro":false,"fullname":"Yuqing Li","user":"MindscapeRAG","type":"user","name":"MindscapeRAG"},"summary":"Relevance is a query-dependent estimate of whether a document or excerpt contains useful evidence. Existing retrieval agents use relevance to select top-k content, but document relevance alone cannot localize, compose, or verify the evidence required by complex questions. Direct Corpus Interaction (DCI) enables such fine-grained operations through grep-style exploration, but its relevance-agnostic search can expose useful clues late and delay convergence. Recent advances use relevance to narrow the corpus into a working space for interaction. Once interaction begins, however, relevance still does not directly guide which documents grep searches first or distinguish informative excerpts from a broad set of matches to let LLMs see them first. We introduce the Relevance-Aware RipGrep Search Agent (RARG), which turns relevance into an execution prior for corpus interaction. RARG provides coarse-to-fine relevance guidance: it orders documents for sequential 'ripgrep' traversal to expose globally relevant clues earlier, initializes promising entry points with query-relevant paragraphs, and reranks grep matches to surface informative excerpts that document-level ranking may otherwise obscure. Across challenging browse question answering and reasoning-intensive retrieval, RARG improves the accuracy--efficiency frontier over retrieval-based and direct-interaction agents. These results demonstrate that relevance-aware interaction enables faster and more reliable search convergence.","upvotes":70,"discussionId":"6a69694a9d3a1231d492b85e","projectPage":"https://qdcassie-li.github.io/RARG/","githubRepo":"https://github.com/LeqsNaN/RARG","githubRepoAddedBy":"user","githubStars":23,"organization":{"_id":"66543b6e420092799d2f625c","name":"tencent","fullname":"Tencent","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/5dd96eb166059660ed1ee413/Lp3m-XLpjQGwBItlvn69q.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6891bbff78946201296b4592","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6891bbff78946201296b4592/ECmWBrlfeonPg0HzmQ_sW.png","isPro":false,"fullname":"Yuqing Li","user":"MindscapeRAG","type":"user"},{"_id":"6670f52a7a0446ec70e11c96","avatarUrl":"/avatars/7786b9d947d7bbaf95013feee3a3bf39.svg","isPro":false,"fullname":"Jiangnan Li","user":"lossisnotanumber","type":"user"},{"_id":"69529466bf5239d747e1ce35","avatarUrl":"/avatars/fa851cafb4a577488151710b0a29c578.svg","isPro":false,"fullname":"ye","user":"nonoonthebeat","type":"user"},{"_id":"60ab6b2ee3de7c7440abb845","avatarUrl":"/avatars/22916bece3b5b951c016bf2ddd8dda1c.svg","isPro":false,"fullname":"Cindy","user":"ttchungc","type":"user"},{"_id":"68a04eef7100a16137e6d07c","avatarUrl":"/avatars/16f33c6dd39daeedb5a8635828d39610.svg","isPro":false,"fullname":"HAO","user":"calmfoxo","type":"user"},{"_id":"67af92045a86287292026808","avatarUrl":"/avatars/8bad9272fe73ba04e077b5484837c8d3.svg","isPro":false,"fullname":"Mo","user":"BishopGorov","type":"user"},{"_id":"689ea38daafdf21ed7c40d34","avatarUrl":"/avatars/fa05f9894bf3139710063e4322727a3f.svg","isPro":false,"fullname":"KN","user":"KaNaaaRL","type":"user"},{"_id":"67cdb9373ba99d1c0cd8209e","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/IZlY6aKMb2T44T55LShTr.png","isPro":false,"fullname":"makescraps","user":"makescraps","type":"user"},{"_id":"65b7c216842cbb1346a8b292","avatarUrl":"/avatars/62bfe59d54d12313cd467aec293c6a7f.svg","isPro":false,"fullname":"frank","user":"joefy","type":"user"},{"_id":"6401a692f98fbc64bcd65f40","avatarUrl":"/avatars/75cfed4422195bf3963b41781052d430.svg","isPro":false,"fullname":"huazhang","user":"huazhang","type":"user"},{"_id":"688f7f48315a619cea7f164f","avatarUrl":"/avatars/e16c52a49db10c8a2beaa7ed81196b17.svg","isPro":false,"fullname":"SituatedEmb","user":"SituatedEmbedding","type":"user"},{"_id":"6a054599891abe09465b58ec","avatarUrl":"/avatars/1d25ffa228408fc837760900fcea39f6.svg","isPro":false,"fullname":"zzli","user":"ZzzzzLi","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":2,"organization":{"_id":"66543b6e420092799d2f625c","name":"tencent","fullname":"Tencent","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/5dd96eb166059660ed1ee413/Lp3m-XLpjQGwBItlvn69q.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.24223.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2607.24223 in a model README.md to link it from this page.

Cite arxiv.org/abs/2607.24223 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2607.24223 in a Space README.md to link it from this page.