SproutRAG is a retrieval-augmented generation stack built for structured, multi-granularity evidence. It combines hierarchical attention-based indexing with flexible retrieval, optional reranking, and generation pipelines. The result is a practical system for building and evaluating RAG workflows end-to-end.</p>\n","updatedAt":"2026-06-22T17:20:43.582Z","author":{"_id":"60e32baedc56466240084155","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/60e32baedc56466240084155/Q-cJh3Q3-vvMbe749Gt5B.jpeg","fullname":"Amirhossein Abaskohi","name":"AmirhosseinAbaskohi","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":4,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8084122538566589},"editors":["AmirhosseinAbaskohi"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/60e32baedc56466240084155/Q-cJh3Q3-vvMbe749Gt5B.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.18381","authors":[{"_id":"6a396e56fdcd3514343bb3e8","name":"Amirhossein Abaskohi","hidden":false},{"_id":"6a396e56fdcd3514343bb3e9","name":"Issam H. Laradji","hidden":false},{"_id":"6a396e56fdcd3514343bb3ea","name":"Peter West","hidden":false},{"_id":"6a396e56fdcd3514343bb3eb","name":"Giuseppe Carenini","hidden":false}],"publishedAt":"2026-06-16T18:28:00.000Z","submittedOnDailyAt":"2026-06-22T00:00:00.000Z","title":"SproutRAG: Attention-Guided Tree Search with Progressive Embeddings for Long-Document RAG","submittedOnDailyBy":{"_id":"60e32baedc56466240084155","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/60e32baedc56466240084155/Q-cJh3Q3-vvMbe749Gt5B.jpeg","isPro":false,"fullname":"Amirhossein Abaskohi","user":"AmirhosseinAbaskohi","type":"user","name":"AmirhosseinAbaskohi"},"summary":"Retrieval-augmented generation (RAG) systems must balance retrieval granularity with contextual coherence, a challenge that existing methods address through LLM-guided chunking, single-level context expansion, or hierarchical summarization. These approaches variously depend on costly LLM calls during indexing or retrieval, limit context aggregation to a single granularity level, or introduce information loss through summarization. We present SproutRAG, an attention-guided hierarchical RAG framework that addresses this trade-off by organizing sentence-level chunks into progressively larger but semantically coherent units, using learned inter-sentence attention to construct a binary chunking tree. Unlike prior approaches that rely on external LLMs, fixed context expansion, or lossy summarization, SproutRAG learns which attention heads and layers best capture semantic document structure, enabling multi-granularity retrieval without additional LLM calls or compressed summaries. At retrieval time, SproutRAG uses hierarchical beam search to retrieve candidates at multiple granularities, capturing multi-sentence relevance beyond flat retrieval. The framework is trained end-to-end with a joint objective that improves both embeddings and tree structure. Experiments across four benchmarks spanning scientific, legal, and open-domain settings demonstrate that SproutRAG improves information efficiency (IE) by 6.1% on average over the strongest baseline. Code is available on https://github.com/AmirAbaskohi/SproutRAG.","upvotes":8,"discussionId":"6a396e56fdcd3514343bb3ec","githubRepo":"https://github.com/AmirAbaskohi/SproutRAG","githubRepoAddedBy":"user","ai_summary":"SproutRAG is an attention-guided hierarchical retrieval-augmented generation framework that organizes sentence-level chunks into semantically coherent units using learned inter-sentence attention, enabling multi-granularity retrieval without additional LLM calls or summarization.","ai_keywords":["retrieval-augmented generation","hierarchical RAG","attention-guided","sentence-level chunks","semantic coherence","inter-sentence attention","binary chunking tree","hierarchical beam search","end-to-end training","joint objective","information efficiency"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":5,"organization":{"_id":"6723d7c5f0e98abd466cd7df","name":"UBC-VisionLanguage","fullname":"Vision and NLP Group at University of British Columbia","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6712caf88a321497c3a15673/0mNkKMu8h74rP2Hzo1O5U.jpeg"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6228ede94323cef93a956b24","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6228ede94323cef93a956b24/b0t3qorPnU-0M_u3dZWqF.jpeg","isPro":false,"fullname":"AmirHossein DabiriAghdam","user":"AmirHossein1378","type":"user"},{"_id":"6a2da6c8ca070ee12c6e396c","avatarUrl":"/avatars/0355287dcabaa67dbc7f0b10b87451f9.svg","isPro":false,"fullname":"Joe Mama","user":"JoeMama123123123","type":"user"},{"_id":"66105ffd174b378a724fb56d","avatarUrl":"/avatars/a32540ce3eb0cc06fe6aa39127dc6ad2.svg","isPro":false,"fullname":"Gaetano Cimino","user":"Gaetano98","type":"user"},{"_id":"60e32baedc56466240084155","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/60e32baedc56466240084155/Q-cJh3Q3-vvMbe749Gt5B.jpeg","isPro":false,"fullname":"Amirhossein Abaskohi","user":"AmirhosseinAbaskohi","type":"user"},{"_id":"6a2718c4ee8c9826408a267d","avatarUrl":"/avatars/75b672abbc4fba71f784503e86753088.svg","isPro":false,"fullname":"SARA VAKILI","user":"saravakili123","type":"user"},{"_id":"6a271c4928e4d750aa0e88cd","avatarUrl":"/avatars/1fd882587298d5a30080b54d1b6c96e5.svg","isPro":false,"fullname":"Asami tanaka","user":"Asami105","type":"user"},{"_id":"64062855692855e65ae31688","avatarUrl":"/avatars/e35d22a037b8b35422d3ee982f133076.svg","isPro":false,"fullname":"Issam Laradji","user":"issamlaradji","type":"user"},{"_id":"607f060442beb4da0f990182","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/607f060442beb4da0f990182/j5W2tLyU6JqkaTf3kv66s.jpeg","isPro":false,"fullname":"Patrice Bechard","user":"patricebechard","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"6723d7c5f0e98abd466cd7df","name":"UBC-VisionLanguage","fullname":"Vision and NLP Group at University of British Columbia","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6712caf88a321497c3a15673/0mNkKMu8h74rP2Hzo1O5U.jpeg"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.18381.md","query":{}}">
SproutRAG: Attention-Guided Tree Search with Progressive Embeddings for Long-Document RAG
Abstract
SproutRAG is an attention-guided hierarchical retrieval-augmented generation framework that organizes sentence-level chunks into semantically coherent units using learned inter-sentence attention, enabling multi-granularity retrieval without additional LLM calls or summarization.
Retrieval-augmented generation (RAG) systems must balance retrieval granularity with contextual coherence, a challenge that existing methods address through LLM-guided chunking, single-level context expansion, or hierarchical summarization. These approaches variously depend on costly LLM calls during indexing or retrieval, limit context aggregation to a single granularity level, or introduce information loss through summarization. We present SproutRAG, an attention-guided hierarchical RAG framework that addresses this trade-off by organizing sentence-level chunks into progressively larger but semantically coherent units, using learned inter-sentence attention to construct a binary chunking tree. Unlike prior approaches that rely on external LLMs, fixed context expansion, or lossy summarization, SproutRAG learns which attention heads and layers best capture semantic document structure, enabling multi-granularity retrieval without additional LLM calls or compressed summaries. At retrieval time, SproutRAG uses hierarchical beam search to retrieve candidates at multiple granularities, capturing multi-sentence relevance beyond flat retrieval. The framework is trained end-to-end with a joint objective that improves both embeddings and tree structure. Experiments across four benchmarks spanning scientific, legal, and open-domain settings demonstrate that SproutRAG improves information efficiency (IE) by 6.1% on average over the strongest baseline. Code is available on https://github.com/AmirAbaskohi/SproutRAG.
Community
SproutRAG is a retrieval-augmented generation stack built for structured, multi-granularity evidence. It combines hierarchical attention-based indexing with flexible retrieval, optional reranking, and generation pipelines. The result is a practical system for building and evaluating RAG workflows end-to-end.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2606.18381 in a model README.md to link it from this page.
Cite arxiv.org/abs/2606.18381 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2606.18381 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.