Large-taxonomy retrieval often assumes that the input already expresses the target concept. In many settings, however, the input is indirect evidence, such as a table cell whose meaning depends on its row, column, datatype, and context. We call this mismatch the retrieval readiness gap. Our analysis shows that the current index retrieves the target reliably when its semantics are explicit, while raw evidence often leaves it deep in the ranking. We propose Factorized Hypothesis Search (FHS), which maintains multiple partial interpretations over named semantic dimensions. These hypotheses support structured query rendering, multi-hypothesis retrieval, and dimension-level candidate verification. On both financial taxonomy tagging and CodiEsp clinical coding tasks, FHS achieves the best Recall@1, MRR, and final accuracy among the non-oracle methods. Replacing the factorized hypothesis path with a free-text ensemble causes the largest drop in head-ranking performance, while sequential refinement provides no additional gain over FHS's strong parallel first round.</p>\n","updatedAt":"2026-08-11T04:21:08.752Z","author":{"_id":"65d76cc5b9b7b8bf88faa916","avatarUrl":"/avatars/d95232cd0c307efab6197ade1a66190b.svg","fullname":"Yan Wang","name":"YanAdjeNole","type":"user","isPro":true,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":7,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8925758600234985},"editors":["YanAdjeNole"],"editorAvatarUrls":["/avatars/d95232cd0c307efab6197ade1a66190b.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2608.06614","authors":[{"_id":"6a7aa2f8019ce76dc7b3aad8","name":"Linhai Ma","hidden":false},{"_id":"6a7aa2f8019ce76dc7b3aad9","name":"Ethan F. Wei","hidden":false},{"_id":"6a7aa2f8019ce76dc7b3aada","name":"Xueqing Peng","hidden":false},{"_id":"6a7aa2f8019ce76dc7b3aadb","user":{"_id":"65d76cc5b9b7b8bf88faa916","avatarUrl":"/avatars/d95232cd0c307efab6197ade1a66190b.svg","isPro":true,"fullname":"Yan Wang","user":"YanAdjeNole","type":"user","name":"YanAdjeNole"},"name":"Yan Wang","status":"claimed_verified","statusLastChangedAt":"2026-08-11T08:45:04.516Z","hidden":false},{"_id":"6a7aa2f8019ce76dc7b3aadc","name":"Lingfei Qian","hidden":false},{"_id":"6a7aa2f8019ce76dc7b3aadd","name":"Víctor Gutiérrez-Basulto","hidden":false}],"publishedAt":"2026-08-06T00:00:00.000Z","submittedOnDailyAt":"2026-08-11T00:00:00.000Z","title":"Factorized Hypothesis Search for Evidence-to-Taxonomy Retrieval","submittedOnDailyBy":{"_id":"65d76cc5b9b7b8bf88faa916","avatarUrl":"/avatars/d95232cd0c307efab6197ade1a66190b.svg","isPro":true,"fullname":"Yan Wang","user":"YanAdjeNole","type":"user","name":"YanAdjeNole"},"summary":"Large-taxonomy retrieval often assumes that the input already expresses the target concept. In many settings, however, the input is indirect evidence, such as a table cell whose meaning depends on its row, column, datatype, and context. We call this mismatch the retrieval readiness gap. Our analysis shows that the current index retrieves the target reliably when its semantics are explicit, while raw evidence often leaves it deep in the ranking. We propose Factorized Hypothesis Search (FHS), which maintains multiple partial interpretations over named semantic dimensions. These hypotheses support structured query rendering, multi-hypothesis retrieval, and dimension-level candidate verification. On both financial taxonomy tagging and CodiEsp clinical coding tasks, FHS achieves the best Recall@1, MRR, and final accuracy among the non-oracle methods. Replacing the factorized hypothesis path with a free-text ensemble causes the largest drop in head-ranking performance, while sequential refinement provides no additional gain over FHS's strong parallel first round.","upvotes":3,"discussionId":"6a7aa2f8019ce76dc7b3aade","ai_summary":"Factorized Hypothesis Search improves retrieval of implicit concepts by maintaining parallel partial interpretations across semantic dimensions for structured querying and verification.","ai_keywords":["Factorized Hypothesis Search","named semantic dimensions","structured query rendering","multi-hypothesis retrieval","dimension-level candidate verification","financial taxonomy tagging","CodiEsp clinical coding"],"ai_summary_model":"thinkingmachines/Inkling-Small","organization":{"_id":"658f4413674349122c0708e9","name":"TheFinAI","fullname":"The Fin AI","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/63b58ed5889aa6707f0bb0f4/ZK5nQKw34W3-eH3p4NAYc.jpeg"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"65d76cc5b9b7b8bf88faa916","avatarUrl":"/avatars/d95232cd0c307efab6197ade1a66190b.svg","isPro":true,"fullname":"Yan Wang","user":"YanAdjeNole","type":"user"},{"_id":"678554189d70a30d526a0daf","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/hv5jjTv7kNx_S8-07KQH9.png","isPro":false,"fullname":"Rinnkai Ba","user":"lm2445","type":"user"},{"_id":"63ac5701c21e60a3e9b58aa7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63ac5701c21e60a3e9b58aa7/g6EX7diOpuA94R2ab-rZC.png","isPro":true,"fullname":"Dipankar Sarkar","user":"dipankarsarkar","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"658f4413674349122c0708e9","name":"TheFinAI","fullname":"The Fin AI","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/63b58ed5889aa6707f0bb0f4/ZK5nQKw34W3-eH3p4NAYc.jpeg"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2608/2608.06614.md","query":{}}">
Factorized Hypothesis Search for Evidence-to-Taxonomy Retrieval
Abstract
Factorized Hypothesis Search improves retrieval of implicit concepts by maintaining parallel partial interpretations across semantic dimensions for structured querying and verification.
Large-taxonomy retrieval often assumes that the input already expresses the target concept. In many settings, however, the input is indirect evidence, such as a table cell whose meaning depends on its row, column, datatype, and context. We call this mismatch the retrieval readiness gap. Our analysis shows that the current index retrieves the target reliably when its semantics are explicit, while raw evidence often leaves it deep in the ranking. We propose Factorized Hypothesis Search (FHS), which maintains multiple partial interpretations over named semantic dimensions. These hypotheses support structured query rendering, multi-hypothesis retrieval, and dimension-level candidate verification. On both financial taxonomy tagging and CodiEsp clinical coding tasks, FHS achieves the best Recall@1, MRR, and final accuracy among the non-oracle methods. Replacing the factorized hypothesis path with a free-text ensemble causes the largest drop in head-ranking performance, while sequential refinement provides no additional gain over FHS's strong parallel first round.
Community
Large-taxonomy retrieval often assumes that the input already expresses the target concept. In many settings, however, the input is indirect evidence, such as a table cell whose meaning depends on its row, column, datatype, and context. We call this mismatch the retrieval readiness gap. Our analysis shows that the current index retrieves the target reliably when its semantics are explicit, while raw evidence often leaves it deep in the ranking. We propose Factorized Hypothesis Search (FHS), which maintains multiple partial interpretations over named semantic dimensions. These hypotheses support structured query rendering, multi-hypothesis retrieval, and dimension-level candidate verification. On both financial taxonomy tagging and CodiEsp clinical coding tasks, FHS achieves the best Recall@1, MRR, and final accuracy among the non-oracle methods. Replacing the factorized hypothesis path with a free-text ensemble causes the largest drop in head-ranking performance, while sequential refinement provides no additional gain over FHS's strong parallel first round.
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/2608.06614 in a model README.md to link it from this page.
Cite arxiv.org/abs/2608.06614 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2608.06614 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.