{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WB44C27LN5FPXLBZLG2DX56IF4","short_pith_number":"pith:WB44C27L","schema_version":"1.0","canonical_sha256":"b079c16beb6f4afbac3959b43bf7c82f1a19abcd306dae7a41f1444c73817928","source":{"kind":"arxiv","id":"2505.15792","version":1},"attestation_state":"computed","paper":{"title":"Long-Form Information Alignment Evaluation Beyond Atomic Facts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Danna Zheng, Jeff Z. Pan, Mirella Lapata","submitted_at":"2025-05-21T17:46:38Z","abstract_excerpt":"Information alignment evaluators are vital for various NLG evaluation tasks and trustworthy LLM deployment, reducing hallucinations and enhancing user trust. Current fine-grained methods, like FactScore, verify facts individually but neglect inter-fact dependencies, enabling subtle vulnerabilities. In this work, we introduce MontageLie, a challenging benchmark that constructs deceptive narratives by \"montaging\" truthful statements without introducing explicit hallucinations. We demonstrate that both coarse-grained LLM-based evaluators and current fine-grained frameworks are susceptible to this"},"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":"2505.15792","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-21T17:46:38Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"01aa2910f5c895ab765960e9b3c80e8aeb1fae920608b75e1aea8c7700f46973","abstract_canon_sha256":"12138f8527c146f6113ceb07df98c85d6cb03cb1ea29ad3bb0bf6e45d00d0e81"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:06:58.101141Z","signature_b64":"hS3Amcd2i3xadM2MD1YO2x0iNTpvLpndlwwTzIwnhNkxS/FoxAKHN0gB+vS1ciyYeMfn3iVkVZVdCdYKEi3NAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b079c16beb6f4afbac3959b43bf7c82f1a19abcd306dae7a41f1444c73817928","last_reissued_at":"2026-07-05T11:06:58.100710Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:06:58.100710Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Long-Form Information Alignment Evaluation Beyond Atomic Facts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Danna Zheng, Jeff Z. Pan, Mirella Lapata","submitted_at":"2025-05-21T17:46:38Z","abstract_excerpt":"Information alignment evaluators are vital for various NLG evaluation tasks and trustworthy LLM deployment, reducing hallucinations and enhancing user trust. Current fine-grained methods, like FactScore, verify facts individually but neglect inter-fact dependencies, enabling subtle vulnerabilities. In this work, we introduce MontageLie, a challenging benchmark that constructs deceptive narratives by \"montaging\" truthful statements without introducing explicit hallucinations. We demonstrate that both coarse-grained LLM-based evaluators and current fine-grained frameworks are susceptible to this"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.15792","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/2505.15792/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":"2505.15792","created_at":"2026-07-05T11:06:58.100766+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.15792v1","created_at":"2026-07-05T11:06:58.100766+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.15792","created_at":"2026-07-05T11:06:58.100766+00:00"},{"alias_kind":"pith_short_12","alias_value":"WB44C27LN5FP","created_at":"2026-07-05T11:06:58.100766+00:00"},{"alias_kind":"pith_short_16","alias_value":"WB44C27LN5FPXLBZ","created_at":"2026-07-05T11:06:58.100766+00:00"},{"alias_kind":"pith_short_8","alias_value":"WB44C27L","created_at":"2026-07-05T11:06:58.100766+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WB44C27LN5FPXLBZLG2DX56IF4","json":"https://pith.science/pith/WB44C27LN5FPXLBZLG2DX56IF4.json","graph_json":"https://pith.science/api/pith-number/WB44C27LN5FPXLBZLG2DX56IF4/graph.json","events_json":"https://pith.science/api/pith-number/WB44C27LN5FPXLBZLG2DX56IF4/events.json","paper":"https://pith.science/paper/WB44C27L"},"agent_actions":{"view_html":"https://pith.science/pith/WB44C27LN5FPXLBZLG2DX56IF4","download_json":"https://pith.science/pith/WB44C27LN5FPXLBZLG2DX56IF4.json","view_paper":"https://pith.science/paper/WB44C27L","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.15792&json=true","fetch_graph":"https://pith.science/api/pith-number/WB44C27LN5FPXLBZLG2DX56IF4/graph.json","fetch_events":"https://pith.science/api/pith-number/WB44C27LN5FPXLBZLG2DX56IF4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WB44C27LN5FPXLBZLG2DX56IF4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WB44C27LN5FPXLBZLG2DX56IF4/action/storage_attestation","attest_author":"https://pith.science/pith/WB44C27LN5FPXLBZLG2DX56IF4/action/author_attestation","sign_citation":"https://pith.science/pith/WB44C27LN5FPXLBZLG2DX56IF4/action/citation_signature","submit_replication":"https://pith.science/pith/WB44C27LN5FPXLBZLG2DX56IF4/action/replication_record"}},"created_at":"2026-07-05T11:06:58.100766+00:00","updated_at":"2026-07-05T11:06:58.100766+00:00"}