{"paper":{"title":"ProCompNav: Proactive Instance Navigation with Comparative Judgment for Ambiguous User Queries","license":"http://creativecommons.org/licenses/by/4.0/","headline":"ProCompNav resolves ambiguous navigation queries by asking binary questions that split candidate pools.","cross_cats":["cs.RO"],"primary_cat":"cs.AI","authors_text":"Hyejin Park, Jungseul Ok, Junhyuk Kwon, Kyle Min, Seungjoon Lee","submitted_at":"2026-05-07T13:19:03Z","abstract_excerpt":"Natural-language instance navigation becomes challenging when the initial user request does not uniquely specify the target instance. A practical agent should reduce the user's burden by actively asking only the information needed to distinguish the target from similar distractors, rather than requiring a detailed description upfront. Existing approaches often fall short of this goal: they may stop at the first plausible candidate before sufficiently exploring alternatives, or, even after collecting multiple candidates, ask about the target's attributes derived from individual candidates rathe"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"On CoIN-Bench, ProCompNav improves Success Rate over interactive baselines with the same minimal input and non-interactive baselines with detailed descriptions, while substantially reducing Response Length; it also achieves state-of-the-art Success Rate on TextNav.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The framework assumes that attribute-value pairs can be reliably extracted from candidates and that binary answers will correctly and completely prune the pool without introducing new errors or requiring follow-up clarification.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"ProCompNav builds a candidate pool from ambiguous queries then uses pool-splitting binary questions for disambiguation, improving success rate and shortening responses on CoIN-Bench and TextNav.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"ProCompNav resolves ambiguous navigation queries by asking binary questions that split candidate pools.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"bf7d63e25f4010573954691b3b1d951d9f6b3319af364f80a86e1444b8a35a1c"},"source":{"id":"2605.06223","kind":"arxiv","version":3},"verdict":{"id":"aab0c92d-00a6-4f11-9a94-cb37597b1af5","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T16:56:40.611075Z","strongest_claim":"On CoIN-Bench, ProCompNav improves Success Rate over interactive baselines with the same minimal input and non-interactive baselines with detailed descriptions, while substantially reducing Response Length; it also achieves state-of-the-art Success Rate on TextNav.","one_line_summary":"ProCompNav builds a candidate pool from ambiguous queries then uses pool-splitting binary questions for disambiguation, improving success rate and shortening responses on CoIN-Bench and TextNav.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The framework assumes that attribute-value pairs can be reliably extracted from candidates and that binary answers will correctly and completely prune the pool without introducing new errors or requiring follow-up clarification.","pith_extraction_headline":"ProCompNav resolves ambiguous navigation queries by asking binary questions that split candidate pools."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.06223/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T19:01:19.118285Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T12:53:36.665498Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"40a7a3b6f12e050f3b039f68c62b78c9677fc0862cf16487da3fd831624bac16"},"references":{"count":49,"sample":[{"doi":"","year":null,"title":"FirstName Alpher , title =","work_id":"42297990-8783-41a1-b0fa-8ccdbf630852","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Journal of Foo , volume = 13, number = 1, pages =","work_id":"65a8b3d0-af84-4f68-87eb-101c85ab18b2","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Journal of Foo , volume = 14, number = 1, pages =","work_id":"b3089947-bd36-4a24-9199-cc535e299537","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"FirstName Alpher and FirstName Gamow , title =","work_id":"caed320b-7cdc-41ca-bb08-00fb14feec62","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Computer Vision -- ECCV 2022 , year =","work_id":"ed75d4d8-3228-4ebf-8e21-c1feb9ce9cae","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":49,"snapshot_sha256":"b19fa386be84b38899124552eff90755e1d4cc41b5979fbac7310a7ae69d3e4b","internal_anchors":4},"formal_canon":{"evidence_count":2,"snapshot_sha256":"8c2f9db37d08160d8bf4f0986d1159da9923633bc6641927efc00802c1758511"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}