{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:LJNLCYJGBDGVHYNMC75CACBHWD","short_pith_number":"pith:LJNLCYJG","schema_version":"1.0","canonical_sha256":"5a5ab1612608cd53e1ac17fa200827b0ed8fa8f801d97a9fe8045d1d0dbe42c3","source":{"kind":"arxiv","id":"2407.16221","version":2},"attestation_state":"computed","paper":{"title":"Do LLMs Know When to NOT Answer? Investigating Abstention Abilities of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Masoud Hashemi, Nishanth Madhusudhan, Sathwik Tejaswi Madhusudhan, Vikas Yadav","submitted_at":"2024-07-23T06:56:54Z","abstract_excerpt":"Abstention Ability (AA) is a critical aspect of Large Language Model (LLM) reliability, referring to an LLM's capability to withhold responses when uncertain or lacking a definitive answer, without compromising performance. Although previous studies have attempted to improve AA, they lack a standardised evaluation method and remain unsuitable for black-box models where token prediction probabilities are inaccessible. This makes comparative analysis challenging, especially for state-of-the-art closed-source commercial LLMs. This paper bridges this gap by introducing a black-box evaluation appro"},"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":"2407.16221","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-23T06:56:54Z","cross_cats_sorted":[],"title_canon_sha256":"175b6e7345324165af029bea5ffa6bbe751d4d77d6ac871c5f1fa5c2863937a7","abstract_canon_sha256":"241fe045c600f9836224c56d8778880ef40ff327cb9514b58978b2fe2998e984"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:10:48.992692Z","signature_b64":"idxZFx66AcVDBEDi31j6/rsrBs862sDD+pcubF4/NauWifpmJR4DSAU9s2qJXsfbvB6m1k2hsjewDaJtHmLLAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a5ab1612608cd53e1ac17fa200827b0ed8fa8f801d97a9fe8045d1d0dbe42c3","last_reissued_at":"2026-07-05T09:10:48.992157Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:10:48.992157Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do LLMs Know When to NOT Answer? Investigating Abstention Abilities of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Masoud Hashemi, Nishanth Madhusudhan, Sathwik Tejaswi Madhusudhan, Vikas Yadav","submitted_at":"2024-07-23T06:56:54Z","abstract_excerpt":"Abstention Ability (AA) is a critical aspect of Large Language Model (LLM) reliability, referring to an LLM's capability to withhold responses when uncertain or lacking a definitive answer, without compromising performance. Although previous studies have attempted to improve AA, they lack a standardised evaluation method and remain unsuitable for black-box models where token prediction probabilities are inaccessible. This makes comparative analysis challenging, especially for state-of-the-art closed-source commercial LLMs. This paper bridges this gap by introducing a black-box evaluation appro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.16221","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/2407.16221/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":"2407.16221","created_at":"2026-07-05T09:10:48.992221+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.16221v2","created_at":"2026-07-05T09:10:48.992221+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.16221","created_at":"2026-07-05T09:10:48.992221+00:00"},{"alias_kind":"pith_short_12","alias_value":"LJNLCYJGBDGV","created_at":"2026-07-05T09:10:48.992221+00:00"},{"alias_kind":"pith_short_16","alias_value":"LJNLCYJGBDGVHYNM","created_at":"2026-07-05T09:10:48.992221+00:00"},{"alias_kind":"pith_short_8","alias_value":"LJNLCYJG","created_at":"2026-07-05T09:10:48.992221+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05985","citing_title":"Auto-DSM Under the Lens: A Black-Box Evaluation Framework for LLM-Based DSM Generation","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"2606.23937","citing_title":"When Retrieval Metrics Mislead: Measuring Policy Signal in Long-Horizon Tool-Use Agents","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2603.22161","citing_title":"Causal Evidence that Language Models use Confidence to Drive Behavior","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03216","citing_title":"BAS: A Decision-Theoretic Approach to Evaluating Large Language Model Confidence","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05379","citing_title":"Partial Evidence Bench: Benchmarking Authorization-Limited Evidence in Agentic Systems","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16752","citing_title":"Don't Start What You Can't Finish: A Counterfactual Audit of Support-State Triage in LLM Agents","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LJNLCYJGBDGVHYNMC75CACBHWD","json":"https://pith.science/pith/LJNLCYJGBDGVHYNMC75CACBHWD.json","graph_json":"https://pith.science/api/pith-number/LJNLCYJGBDGVHYNMC75CACBHWD/graph.json","events_json":"https://pith.science/api/pith-number/LJNLCYJGBDGVHYNMC75CACBHWD/events.json","paper":"https://pith.science/paper/LJNLCYJG"},"agent_actions":{"view_html":"https://pith.science/pith/LJNLCYJGBDGVHYNMC75CACBHWD","download_json":"https://pith.science/pith/LJNLCYJGBDGVHYNMC75CACBHWD.json","view_paper":"https://pith.science/paper/LJNLCYJG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.16221&json=true","fetch_graph":"https://pith.science/api/pith-number/LJNLCYJGBDGVHYNMC75CACBHWD/graph.json","fetch_events":"https://pith.science/api/pith-number/LJNLCYJGBDGVHYNMC75CACBHWD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LJNLCYJGBDGVHYNMC75CACBHWD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LJNLCYJGBDGVHYNMC75CACBHWD/action/storage_attestation","attest_author":"https://pith.science/pith/LJNLCYJGBDGVHYNMC75CACBHWD/action/author_attestation","sign_citation":"https://pith.science/pith/LJNLCYJGBDGVHYNMC75CACBHWD/action/citation_signature","submit_replication":"https://pith.science/pith/LJNLCYJGBDGVHYNMC75CACBHWD/action/replication_record"}},"created_at":"2026-07-05T09:10:48.992221+00:00","updated_at":"2026-07-05T09:10:48.992221+00:00"}