{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MOBSON4TDXY5ZOWUOGK3CSGM67","short_pith_number":"pith:MOBSON4T","schema_version":"1.0","canonical_sha256":"63832737931df1dcbad47195b148ccf7e4a67bec0dea85fa48b8efcf9a766d1b","source":{"kind":"arxiv","id":"2510.16492","version":4},"attestation_state":"computed","paper":{"title":"Check Yourself Before You Wreck Yourself: Selectively Quitting Improves LLM Agent Safety","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Benjamin Plaut, Khanh Nguyen, Ponnurangam Kumaragurum, Vamshi Krishna Bonagiri","submitted_at":"2025-10-18T13:22:19Z","abstract_excerpt":"As Large Language Model (LLM) agents increasingly operate in complex environments with real-world consequences, their safety becomes critical. While uncertainty quantification is well-studied for single-turn tasks, multi-turn agentic scenarios with real-world tool access present unique challenges where uncertainties and ambiguities compound, leading to severe or catastrophic risks beyond traditional text generation failures. We propose using \"quitting\" as a simple yet effective behavioral mechanism for LLM agents to recognize and withdraw from situations where they lack confidence. Leveraging "},"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":"2510.16492","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-10-18T13:22:19Z","cross_cats_sorted":[],"title_canon_sha256":"3291e18f566f1b4502c4d743684530d902615ca7346f66bdb84ab452c28a2ce5","abstract_canon_sha256":"e562c397bca1a99ad8e7e10c8370831f6c79472aec3f0d04c9eb02843a8fa7cd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-29T01:14:24.972433Z","signature_b64":"eB3PevcCxTV/OGBU1IrqNLwX4u61Aa3nV/ixlXXBNFdOd0SpcKtobtus+oO3AVNzkk7ZEhrSQ1mbXzt3QrsEDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"63832737931df1dcbad47195b148ccf7e4a67bec0dea85fa48b8efcf9a766d1b","last_reissued_at":"2026-06-29T01:14:24.971864Z","signature_status":"signed_v1","first_computed_at":"2026-06-29T01:14:24.971864Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Check Yourself Before You Wreck Yourself: Selectively Quitting Improves LLM Agent Safety","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Benjamin Plaut, Khanh Nguyen, Ponnurangam Kumaragurum, Vamshi Krishna Bonagiri","submitted_at":"2025-10-18T13:22:19Z","abstract_excerpt":"As Large Language Model (LLM) agents increasingly operate in complex environments with real-world consequences, their safety becomes critical. While uncertainty quantification is well-studied for single-turn tasks, multi-turn agentic scenarios with real-world tool access present unique challenges where uncertainties and ambiguities compound, leading to severe or catastrophic risks beyond traditional text generation failures. We propose using \"quitting\" as a simple yet effective behavioral mechanism for LLM agents to recognize and withdraw from situations where they lack confidence. Leveraging "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.16492","kind":"arxiv","version":4},"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/2510.16492/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":"2510.16492","created_at":"2026-06-29T01:14:24.971928+00:00"},{"alias_kind":"arxiv_version","alias_value":"2510.16492v4","created_at":"2026-06-29T01:14:24.971928+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.16492","created_at":"2026-06-29T01:14:24.971928+00:00"},{"alias_kind":"pith_short_12","alias_value":"MOBSON4TDXY5","created_at":"2026-06-29T01:14:24.971928+00:00"},{"alias_kind":"pith_short_16","alias_value":"MOBSON4TDXY5ZOWU","created_at":"2026-06-29T01:14:24.971928+00:00"},{"alias_kind":"pith_short_8","alias_value":"MOBSON4T","created_at":"2026-06-29T01:14:24.971928+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2606.28733","citing_title":"Agentic Abstention: Do Agents Know When to Stop Instead of Act?","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2604.15367","citing_title":"SoK: Security of Autonomous LLM Agents in Agentic Commerce","ref_index":98,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MOBSON4TDXY5ZOWUOGK3CSGM67","json":"https://pith.science/pith/MOBSON4TDXY5ZOWUOGK3CSGM67.json","graph_json":"https://pith.science/api/pith-number/MOBSON4TDXY5ZOWUOGK3CSGM67/graph.json","events_json":"https://pith.science/api/pith-number/MOBSON4TDXY5ZOWUOGK3CSGM67/events.json","paper":"https://pith.science/paper/MOBSON4T"},"agent_actions":{"view_html":"https://pith.science/pith/MOBSON4TDXY5ZOWUOGK3CSGM67","download_json":"https://pith.science/pith/MOBSON4TDXY5ZOWUOGK3CSGM67.json","view_paper":"https://pith.science/paper/MOBSON4T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2510.16492&json=true","fetch_graph":"https://pith.science/api/pith-number/MOBSON4TDXY5ZOWUOGK3CSGM67/graph.json","fetch_events":"https://pith.science/api/pith-number/MOBSON4TDXY5ZOWUOGK3CSGM67/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MOBSON4TDXY5ZOWUOGK3CSGM67/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MOBSON4TDXY5ZOWUOGK3CSGM67/action/storage_attestation","attest_author":"https://pith.science/pith/MOBSON4TDXY5ZOWUOGK3CSGM67/action/author_attestation","sign_citation":"https://pith.science/pith/MOBSON4TDXY5ZOWUOGK3CSGM67/action/citation_signature","submit_replication":"https://pith.science/pith/MOBSON4TDXY5ZOWUOGK3CSGM67/action/replication_record"}},"created_at":"2026-06-29T01:14:24.971928+00:00","updated_at":"2026-06-29T01:14:24.971928+00:00"}