DAR introduces an agentic setup where LLMs query statutes on demand through tools rather than receiving all rules in one prompt, showing that this can improve frontier models’ deontic reasoning but often hurts weaker models while greatly increasing token use.</p>\n","updatedAt":"2026-06-04T14:24:17.924Z","author":{"_id":"65cdb4fa6f41d775fe4853f7","avatarUrl":"/avatars/7cd61ae44887ff917a1954ee09466fdf.svg","fullname":"Guangyao Dou","name":"gydou","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":2,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9113203287124634},"editors":["gydou"],"editorAvatarUrls":["/avatars/7cd61ae44887ff917a1954ee09466fdf.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.05009","authors":[{"_id":"6a2189fa3490a593e87b0f0f","name":"Guangyao Dou","hidden":false},{"_id":"6a2189fa3490a593e87b0f10","name":"William Jurayj","hidden":false},{"_id":"6a2189fa3490a593e87b0f11","name":"Nils Holzenberger","hidden":false},{"_id":"6a2189fa3490a593e87b0f12","name":"Benjamin Van Durme","hidden":false}],"publishedAt":"2026-06-03T00:00:00.000Z","submittedOnDailyAt":"2026-06-04T00:00:00.000Z","title":"DAR: Deontic Reasoning with Agentic Harnesses","submittedOnDailyBy":{"_id":"65cdb4fa6f41d775fe4853f7","avatarUrl":"/avatars/7cd61ae44887ff917a1954ee09466fdf.svg","isPro":false,"fullname":"Guangyao Dou","user":"gydou","type":"user","name":"gydou"},"summary":"Deontic reasoning is the task of answering questions by applying explicit rules and policies to case-specific facts, for example computing tax liability under a statute or determining the outcome of an immigration appeal. A key technical challenge for LLM-based deontic reasoning is that the relevant ruleset can be long and cross-referenced, so models may still fail to locate the rules needed for a particular reasoning step. We introduce Deontic Agentic Reasoning (DAR), an agentic reasoning setup in which the model interacts with the statutes on demand. We evaluate DAR under multiple harnesses on hard subsets of DeonticBench. Across these settings, we find that agentic harnesses can push the frontier on deontic reasoning tasks, but improvements are not uniform: weaker models often degrade on numerical tasks while consuming far more tokens.","upvotes":3,"discussionId":"6a2189fa3490a593e87b0f13","projectPage":"https://guangyaodou.github.io/harbor-deonticbench/","githubRepo":"https://github.com/guangyaodou/harbor-deonticbench","githubRepoAddedBy":"user","ai_summary":"Deontic reasoning tasks require applying complex rules and policies, and an agentic approach enables models to dynamically access statutes, showing mixed performance improvements across different model strengths.","ai_keywords":["deontic reasoning","agentic reasoning","statutes","DeonticBench","rule-based reasoning","policy application"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":1,"organization":{"_id":"6137aeeaf8e9dca6e152bccf","name":"jhu-clsp","fullname":"Center for Language and Speech Processing @ JHU","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1631039662102-6137ad94501f80a6f6e1eac9.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"65025370b6595dc45c397340","avatarUrl":"/avatars/9469599b176034548042922c0afa7051.svg","isPro":false,"fullname":"J C","user":"dark-pen","type":"user"},{"_id":"5f6540c65e78cc6b0ed3199d","avatarUrl":"/avatars/0280d4df417855965a0964d22766c012.svg","isPro":false,"fullname":"Daniel Khashabi","user":"danyaljj","type":"user"},{"_id":"644938fcd15756ed2117b7bb","avatarUrl":"/avatars/bc2aab74168c09121cf31d38af4c5b87.svg","isPro":false,"fullname":"Jonathan Ivey","user":"Jonathan-Ivey","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"6137aeeaf8e9dca6e152bccf","name":"jhu-clsp","fullname":"Center for Language and Speech Processing @ JHU","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1631039662102-6137ad94501f80a6f6e1eac9.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.05009.md"}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2606.05009 in a model README.md to link it from this page.

Cite arxiv.org/abs/2606.05009 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2606.05009 in a Space README.md to link it from this page.

Add this paper to a collection to link it from this page.