{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:OQZP7MLEXOUKJJLTX52AJ6YC4K","short_pith_number":"pith:OQZP7MLE","schema_version":"1.0","canonical_sha256":"7432ffb164bba8a4a573bf7404fb02e2bea3b13a0961a2e4192a766615b56699","source":{"kind":"arxiv","id":"2506.11440","version":1},"attestation_state":"computed","paper":{"title":"AbsenceBench: Language Models Can't Tell What's Missing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ari Holtzman, Aryan Shrivastava, Chenhao Tan, Harvey Yiyun Fu, Jared Moore, Peter West","submitted_at":"2025-06-13T03:38:29Z","abstract_excerpt":"Large language models (LLMs) are increasingly capable of processing long inputs and locating specific information within them, as evidenced by their performance on the Needle in a Haystack (NIAH) test. However, while models excel at recalling surprising information, they still struggle to identify clearly omitted information. We introduce AbsenceBench to assesses LLMs' capacity to detect missing information across three domains: numerical sequences, poetry, and GitHub pull requests. AbsenceBench asks models to identify which pieces of a document were deliberately removed, given access to both "},"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":"2506.11440","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-13T03:38:29Z","cross_cats_sorted":[],"title_canon_sha256":"198de4067ead7eb36d48e7b5fe3a6a602d0aea57e3a4fd423347470f9e471725","abstract_canon_sha256":"00219fb9fd545d9dfffe100f2929bf9d4bdacb739b3ce43a248bb721670f2257"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:04.546062Z","signature_b64":"0f5HD4ZDXCPFHe3BRctZ4oWDkH/y5eLKDItbpwicry8Te5Ga1u4TOhu1wUpckZEbwZ3mtxTLjOSdz7eRRYvUAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7432ffb164bba8a4a573bf7404fb02e2bea3b13a0961a2e4192a766615b56699","last_reissued_at":"2026-07-05T11:21:04.545557Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:04.545557Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AbsenceBench: Language Models Can't Tell What's Missing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ari Holtzman, Aryan Shrivastava, Chenhao Tan, Harvey Yiyun Fu, Jared Moore, Peter West","submitted_at":"2025-06-13T03:38:29Z","abstract_excerpt":"Large language models (LLMs) are increasingly capable of processing long inputs and locating specific information within them, as evidenced by their performance on the Needle in a Haystack (NIAH) test. However, while models excel at recalling surprising information, they still struggle to identify clearly omitted information. We introduce AbsenceBench to assesses LLMs' capacity to detect missing information across three domains: numerical sequences, poetry, and GitHub pull requests. AbsenceBench asks models to identify which pieces of a document were deliberately removed, given access to both "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.11440","kind":"arxiv","version":1},"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/2506.11440/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":"2506.11440","created_at":"2026-07-05T11:21:04.545618+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.11440v1","created_at":"2026-07-05T11:21:04.545618+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.11440","created_at":"2026-07-05T11:21:04.545618+00:00"},{"alias_kind":"pith_short_12","alias_value":"OQZP7MLEXOUK","created_at":"2026-07-05T11:21:04.545618+00:00"},{"alias_kind":"pith_short_16","alias_value":"OQZP7MLEXOUKJJLT","created_at":"2026-07-05T11:21:04.545618+00:00"},{"alias_kind":"pith_short_8","alias_value":"OQZP7MLE","created_at":"2026-07-05T11:21:04.545618+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26529","citing_title":"The inattentional gap in task conditioned AI models that omit otherwise reportable safety critical signals","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02680","citing_title":"Locality Does Not Imply Reachability: Boundary Repair in Block-Sparse Causal Attention","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.26145","citing_title":"Ceci n'est pas une explication: Evaluating Explanation Failures as Explainability Pitfalls in Language Learning Systems","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2512.03053","citing_title":"Mitigating hallucinations and omissions in LLMs for invertible problems: An application to hardware logic design automation","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2604.26145","citing_title":"Ceci n'est pas une explication: Evaluating Explanation Failures as Explainability Pitfalls in Language Learning Systems","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OQZP7MLEXOUKJJLTX52AJ6YC4K","json":"https://pith.science/pith/OQZP7MLEXOUKJJLTX52AJ6YC4K.json","graph_json":"https://pith.science/api/pith-number/OQZP7MLEXOUKJJLTX52AJ6YC4K/graph.json","events_json":"https://pith.science/api/pith-number/OQZP7MLEXOUKJJLTX52AJ6YC4K/events.json","paper":"https://pith.science/paper/OQZP7MLE"},"agent_actions":{"view_html":"https://pith.science/pith/OQZP7MLEXOUKJJLTX52AJ6YC4K","download_json":"https://pith.science/pith/OQZP7MLEXOUKJJLTX52AJ6YC4K.json","view_paper":"https://pith.science/paper/OQZP7MLE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.11440&json=true","fetch_graph":"https://pith.science/api/pith-number/OQZP7MLEXOUKJJLTX52AJ6YC4K/graph.json","fetch_events":"https://pith.science/api/pith-number/OQZP7MLEXOUKJJLTX52AJ6YC4K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OQZP7MLEXOUKJJLTX52AJ6YC4K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OQZP7MLEXOUKJJLTX52AJ6YC4K/action/storage_attestation","attest_author":"https://pith.science/pith/OQZP7MLEXOUKJJLTX52AJ6YC4K/action/author_attestation","sign_citation":"https://pith.science/pith/OQZP7MLEXOUKJJLTX52AJ6YC4K/action/citation_signature","submit_replication":"https://pith.science/pith/OQZP7MLEXOUKJJLTX52AJ6YC4K/action/replication_record"}},"created_at":"2026-07-05T11:21:04.545618+00:00","updated_at":"2026-07-05T11:21:04.545618+00:00"}