{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:WS5JKJ7HYW3T5KHLOAPTR47WPJ","short_pith_number":"pith:WS5JKJ7H","schema_version":"1.0","canonical_sha256":"b4ba9527e7c5b73ea8eb701f38f3f67a748f957cfc5b9565470d168aa9f38a63","source":{"kind":"arxiv","id":"2307.01928","version":2},"attestation_state":"computed","paper":{"title":"Robots That Ask For Help: Uncertainty Alignment for Large Language Model Planners","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.AP"],"primary_cat":"cs.RO","authors_text":"Alexandra Bodrova, Allen Z. Ren, Andy Zeng, Anirudha Majumdar, Anushri Dixit, Dorsa Sadigh, Fei Xia, Jake Varley, Leila Takayama, Noah Brown, Peng Xu, Stephen Tu, Sumeet Singh, Zhenjia Xu","submitted_at":"2023-07-04T21:25:12Z","abstract_excerpt":"Large language models (LLMs) exhibit a wide range of promising capabilities -- from step-by-step planning to commonsense reasoning -- that may provide utility for robots, but remain prone to confidently hallucinated predictions. In this work, we present KnowNo, which is a framework for measuring and aligning the uncertainty of LLM-based planners such that they know when they don't know and ask for help when needed. KnowNo builds on the theory of conformal prediction to provide statistical guarantees on task completion while minimizing human help in complex multi-step planning settings. Experim"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2307.01928","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2023-07-04T21:25:12Z","cross_cats_sorted":["cs.AI","stat.AP"],"title_canon_sha256":"78454ea4661971fae8670467ce61f1f98df4fc2392854695c4c5065cd06a02f6","abstract_canon_sha256":"8587dded840f93e04267658e2d0428f10e93628058230cabf647b66d012ccf8f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:47:24.631004Z","signature_b64":"5zsIcKxAvexBecwcShOYlg1N5CAku4O2LEyl4Fi1tz9Kal957U4tP0UVc4tdLo3PHyGeJL0xJb4TfanOFgymBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b4ba9527e7c5b73ea8eb701f38f3f67a748f957cfc5b9565470d168aa9f38a63","last_reissued_at":"2026-07-05T06:47:24.630327Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:47:24.630327Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Robots That Ask For Help: Uncertainty Alignment for Large Language Model Planners","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.AP"],"primary_cat":"cs.RO","authors_text":"Alexandra Bodrova, Allen Z. Ren, Andy Zeng, Anirudha Majumdar, Anushri Dixit, Dorsa Sadigh, Fei Xia, Jake Varley, Leila Takayama, Noah Brown, Peng Xu, Stephen Tu, Sumeet Singh, Zhenjia Xu","submitted_at":"2023-07-04T21:25:12Z","abstract_excerpt":"Large language models (LLMs) exhibit a wide range of promising capabilities -- from step-by-step planning to commonsense reasoning -- that may provide utility for robots, but remain prone to confidently hallucinated predictions. In this work, we present KnowNo, which is a framework for measuring and aligning the uncertainty of LLM-based planners such that they know when they don't know and ask for help when needed. KnowNo builds on the theory of conformal prediction to provide statistical guarantees on task completion while minimizing human help in complex multi-step planning settings. Experim"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.01928","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2307.01928/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2307.01928","created_at":"2026-07-05T06:47:24.630402+00:00"},{"alias_kind":"arxiv_version","alias_value":"2307.01928v2","created_at":"2026-07-05T06:47:24.630402+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.01928","created_at":"2026-07-05T06:47:24.630402+00:00"},{"alias_kind":"pith_short_12","alias_value":"WS5JKJ7HYW3T","created_at":"2026-07-05T06:47:24.630402+00:00"},{"alias_kind":"pith_short_16","alias_value":"WS5JKJ7HYW3T5KHL","created_at":"2026-07-05T06:47:24.630402+00:00"},{"alias_kind":"pith_short_8","alias_value":"WS5JKJ7H","created_at":"2026-07-05T06:47:24.630402+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":14,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21399","citing_title":"Calibration Is Not Control: Why LLM-Agent Oversight Needs Intervention","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18189","citing_title":"Beyond Failure Recovery: An Engagement-Aware Human-in-the-loop Framework for Robotic Systems","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12587","citing_title":"Strategic Decision Support for AI Agents","ref_index":71,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20662","citing_title":"Confidence Laundering in Agent Systems: Why Uncertainty Needs a Latent Carrier","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14358","citing_title":"Uncovering the Representation Geometry of Minimal Cores in Overcomplete Reasoning Traces","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29654","citing_title":"Budgeted Act-or-Defer Multi-Agent LLM Deliberation with Local Reliability Bounds","ref_index":82,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18109","citing_title":"TaskGround: Structured Executable Task Inference for Full-Scene Household Reasoning","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18747","citing_title":"Code as Agent Harness","ref_index":111,"is_internal_anchor":false},{"citing_arxiv_id":"2510.11954","citing_title":"VizCopilot: Fostering Appropriate Reliance on Enterprise Chatbots with Context Visualization","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2601.05991","citing_title":"3D Instruction Ambiguity Detection","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2602.04129","citing_title":"KGLAMP: Knowledge Graph-guided Language model for Adaptive Multi-robot Planning and Replanning","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2304.11477","citing_title":"LLM+P: Empowering Large Language Models with Optimal Planning Proficiency","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27914","citing_title":"Geometry-Calibrated Conformal Abstention for Language Models","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06116","citing_title":"Policy-Guided Stepwise Model Routing for Cost-Effective Reasoning","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WS5JKJ7HYW3T5KHLOAPTR47WPJ","json":"https://pith.science/pith/WS5JKJ7HYW3T5KHLOAPTR47WPJ.json","graph_json":"https://pith.science/api/pith-number/WS5JKJ7HYW3T5KHLOAPTR47WPJ/graph.json","events_json":"https://pith.science/api/pith-number/WS5JKJ7HYW3T5KHLOAPTR47WPJ/events.json","paper":"https://pith.science/paper/WS5JKJ7H"},"agent_actions":{"view_html":"https://pith.science/pith/WS5JKJ7HYW3T5KHLOAPTR47WPJ","download_json":"https://pith.science/pith/WS5JKJ7HYW3T5KHLOAPTR47WPJ.json","view_paper":"https://pith.science/paper/WS5JKJ7H","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2307.01928&json=true","fetch_graph":"https://pith.science/api/pith-number/WS5JKJ7HYW3T5KHLOAPTR47WPJ/graph.json","fetch_events":"https://pith.science/api/pith-number/WS5JKJ7HYW3T5KHLOAPTR47WPJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WS5JKJ7HYW3T5KHLOAPTR47WPJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WS5JKJ7HYW3T5KHLOAPTR47WPJ/action/storage_attestation","attest_author":"https://pith.science/pith/WS5JKJ7HYW3T5KHLOAPTR47WPJ/action/author_attestation","sign_citation":"https://pith.science/pith/WS5JKJ7HYW3T5KHLOAPTR47WPJ/action/citation_signature","submit_replication":"https://pith.science/pith/WS5JKJ7HYW3T5KHLOAPTR47WPJ/action/replication_record"}},"created_at":"2026-07-05T06:47:24.630402+00:00","updated_at":"2026-07-05T06:47:24.630402+00:00"}