{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:GW7J7RCDBFZAV26ZCMCCFDGVL5","short_pith_number":"pith:GW7J7RCD","canonical_record":{"source":{"id":"2504.18413","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-04-25T15:14:25Z","cross_cats_sorted":[],"title_canon_sha256":"f6de3ccdf0d63b3fcf3c24c5dbdc78202cfef7663c7569bd786a1a376552a61d","abstract_canon_sha256":"2bf254d9315def322a760af0add7abb455cd2282f64c5e25745608871cb18b17"},"schema_version":"1.0"},"canonical_sha256":"35be9fc44309720aebd91304228cd55f7225a9d4dda3cbca585511dab9b45b72","source":{"kind":"arxiv","id":"2504.18413","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.18413","created_at":"2026-07-05T10:54:06Z"},{"alias_kind":"arxiv_version","alias_value":"2504.18413v1","created_at":"2026-07-05T10:54:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.18413","created_at":"2026-07-05T10:54:06Z"},{"alias_kind":"pith_short_12","alias_value":"GW7J7RCDBFZA","created_at":"2026-07-05T10:54:06Z"},{"alias_kind":"pith_short_16","alias_value":"GW7J7RCDBFZAV26Z","created_at":"2026-07-05T10:54:06Z"},{"alias_kind":"pith_short_8","alias_value":"GW7J7RCD","created_at":"2026-07-05T10:54:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:GW7J7RCDBFZAV26ZCMCCFDGVL5","target":"record","payload":{"canonical_record":{"source":{"id":"2504.18413","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-04-25T15:14:25Z","cross_cats_sorted":[],"title_canon_sha256":"f6de3ccdf0d63b3fcf3c24c5dbdc78202cfef7663c7569bd786a1a376552a61d","abstract_canon_sha256":"2bf254d9315def322a760af0add7abb455cd2282f64c5e25745608871cb18b17"},"schema_version":"1.0"},"canonical_sha256":"35be9fc44309720aebd91304228cd55f7225a9d4dda3cbca585511dab9b45b72","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:54:06.617291Z","signature_b64":"Nq+ErlFgRsaXEK7zBHE9L8J/+xazcbBxCsz1Y1A8d7VABkmzlhDHtCuUzFGLYYthbGRD6ERZqai8TmSQGZdQBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"35be9fc44309720aebd91304228cd55f7225a9d4dda3cbca585511dab9b45b72","last_reissued_at":"2026-07-05T10:54:06.616810Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:54:06.616810Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2504.18413","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:54:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CyY322PZlXkntgVpy04QA9XEHPEKtLHtXWpts87dvFQbSnpfqrnfQEaAwB9Fjmc1a1TN4hcVJ0CZ6lpcvIDRAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T16:40:13.340043Z"},"content_sha256":"d0b139624b1268c8d5314622f310a9bf50afd202005b6a0be035bd2b2140edf0","schema_version":"1.0","event_id":"sha256:d0b139624b1268c8d5314622f310a9bf50afd202005b6a0be035bd2b2140edf0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:GW7J7RCDBFZAV26ZCMCCFDGVL5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"An Empirical Study of Evaluating Long-form Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Jiafeng Guo, Maarten de Rijke, Ning Xian, Ruqing Zhang, Yixing Fan","submitted_at":"2025-04-25T15:14:25Z","abstract_excerpt":"\\Ac{LFQA} aims to generate lengthy answers to complex questions. This scenario presents great flexibility as well as significant challenges for evaluation. Most evaluations rely on deterministic metrics that depend on string or n-gram matching, while the reliability of large language model-based evaluations for long-form answers remains relatively unexplored. We address this gap by conducting an in-depth study of long-form answer evaluation with the following research questions: (i) To what extent do existing automatic evaluation metrics serve as a substitute for human evaluations? (ii) What a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.18413","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/2504.18413/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:54:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Qzsyj8LUfPWnvPfBuWKBGSTFFLQE0KMTSl43NulgfjEnccJd042H2dnxE+S1AnN5q3/70ag9gLVUD+BDY8F8Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T16:40:13.340760Z"},"content_sha256":"04d114e51c34d288b79de13b3768de6d5c8bbf16c211202c0f52628ddcac169d","schema_version":"1.0","event_id":"sha256:04d114e51c34d288b79de13b3768de6d5c8bbf16c211202c0f52628ddcac169d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GW7J7RCDBFZAV26ZCMCCFDGVL5/bundle.json","state_url":"https://pith.science/pith/GW7J7RCDBFZAV26ZCMCCFDGVL5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GW7J7RCDBFZAV26ZCMCCFDGVL5/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-16T16:40:13Z","links":{"resolver":"https://pith.science/pith/GW7J7RCDBFZAV26ZCMCCFDGVL5","bundle":"https://pith.science/pith/GW7J7RCDBFZAV26ZCMCCFDGVL5/bundle.json","state":"https://pith.science/pith/GW7J7RCDBFZAV26ZCMCCFDGVL5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GW7J7RCDBFZAV26ZCMCCFDGVL5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:GW7J7RCDBFZAV26ZCMCCFDGVL5","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"2bf254d9315def322a760af0add7abb455cd2282f64c5e25745608871cb18b17","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-04-25T15:14:25Z","title_canon_sha256":"f6de3ccdf0d63b3fcf3c24c5dbdc78202cfef7663c7569bd786a1a376552a61d"},"schema_version":"1.0","source":{"id":"2504.18413","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.18413","created_at":"2026-07-05T10:54:06Z"},{"alias_kind":"arxiv_version","alias_value":"2504.18413v1","created_at":"2026-07-05T10:54:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.18413","created_at":"2026-07-05T10:54:06Z"},{"alias_kind":"pith_short_12","alias_value":"GW7J7RCDBFZA","created_at":"2026-07-05T10:54:06Z"},{"alias_kind":"pith_short_16","alias_value":"GW7J7RCDBFZAV26Z","created_at":"2026-07-05T10:54:06Z"},{"alias_kind":"pith_short_8","alias_value":"GW7J7RCD","created_at":"2026-07-05T10:54:06Z"}],"graph_snapshots":[{"event_id":"sha256:04d114e51c34d288b79de13b3768de6d5c8bbf16c211202c0f52628ddcac169d","target":"graph","created_at":"2026-07-05T10:54:06Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2504.18413/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"\\Ac{LFQA} aims to generate lengthy answers to complex questions. This scenario presents great flexibility as well as significant challenges for evaluation. Most evaluations rely on deterministic metrics that depend on string or n-gram matching, while the reliability of large language model-based evaluations for long-form answers remains relatively unexplored. We address this gap by conducting an in-depth study of long-form answer evaluation with the following research questions: (i) To what extent do existing automatic evaluation metrics serve as a substitute for human evaluations? (ii) What a","authors_text":"Jiafeng Guo, Maarten de Rijke, Ning Xian, Ruqing Zhang, Yixing Fan","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-04-25T15:14:25Z","title":"An Empirical Study of Evaluating Long-form Question Answering"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.18413","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:d0b139624b1268c8d5314622f310a9bf50afd202005b6a0be035bd2b2140edf0","target":"record","created_at":"2026-07-05T10:54:06Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"2bf254d9315def322a760af0add7abb455cd2282f64c5e25745608871cb18b17","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-04-25T15:14:25Z","title_canon_sha256":"f6de3ccdf0d63b3fcf3c24c5dbdc78202cfef7663c7569bd786a1a376552a61d"},"schema_version":"1.0","source":{"id":"2504.18413","kind":"arxiv","version":1}},"canonical_sha256":"35be9fc44309720aebd91304228cd55f7225a9d4dda3cbca585511dab9b45b72","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"35be9fc44309720aebd91304228cd55f7225a9d4dda3cbca585511dab9b45b72","first_computed_at":"2026-07-05T10:54:06.616810Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:54:06.616810Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Nq+ErlFgRsaXEK7zBHE9L8J/+xazcbBxCsz1Y1A8d7VABkmzlhDHtCuUzFGLYYthbGRD6ERZqai8TmSQGZdQBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:54:06.617291Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.18413","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d0b139624b1268c8d5314622f310a9bf50afd202005b6a0be035bd2b2140edf0","sha256:04d114e51c34d288b79de13b3768de6d5c8bbf16c211202c0f52628ddcac169d"],"state_sha256":"6be532f87545d8088f5cac0e5b84effc1de28804d66831a49e2e1f09140ec837"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wrhXQE9n6tpCCMG+CePAQMf49yBnhUEkI1EbI+CJe6ilQ0jYErUbgQJb2YA8vkPPwo2MyxxHrHobMfN7CiTfBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-16T16:40:13.348103Z","bundle_sha256":"956e802bb381b4c79ab5ca136b4ac83cc71103b94f9ab4d53e70d783e70c23d1"}}