{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:G4TL5LADVM5EJCPRJCYFRSL6I4","short_pith_number":"pith:G4TL5LAD","schema_version":"1.0","canonical_sha256":"3726beac03ab3a4489f148b058c97e473dfa7dbaf53d5168a17e2e67761d05ea","source":{"kind":"arxiv","id":"2305.18201","version":1},"attestation_state":"computed","paper":{"title":"A Critical Evaluation of Evaluations for Long-form Question Answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Eunsol Choi, Fangyuan Xu, Mohit Iyyer, Yixiao Song","submitted_at":"2023-05-29T16:54:24Z","abstract_excerpt":"Long-form question answering (LFQA) enables answering a wide range of questions, but its flexibility poses enormous challenges for evaluation. We perform the first targeted study of the evaluation of long-form answers, covering both human and automatic evaluation practices. We hire domain experts in seven areas to provide preference judgments over pairs of answers, along with free-form justifications for their choices. We present a careful analysis of experts' evaluation, which focuses on new aspects such as the comprehensiveness of the answer. Next, we examine automatic text generation metric"},"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":"2305.18201","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-29T16:54:24Z","cross_cats_sorted":[],"title_canon_sha256":"f388ce3f587b20437ecfd4b9b05715257732fc2ca03b8fd3180b013b3e6359d4","abstract_canon_sha256":"1cbb25879833a8f1678e2f5fc28b6a8c2620ab310884730fb2c0e0c7c3487bb7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:14:53.390208Z","signature_b64":"TAidClqWemjyVchzHuP7Dix8FTlx0reP1ntar3sPvZ8/HPAj/yVvFrRajM1e3PTvtjoi//P66EXruCw2OnHWCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3726beac03ab3a4489f148b058c97e473dfa7dbaf53d5168a17e2e67761d05ea","last_reissued_at":"2026-07-05T06:14:53.389743Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:14:53.389743Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Critical Evaluation of Evaluations for Long-form Question Answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Eunsol Choi, Fangyuan Xu, Mohit Iyyer, Yixiao Song","submitted_at":"2023-05-29T16:54:24Z","abstract_excerpt":"Long-form question answering (LFQA) enables answering a wide range of questions, but its flexibility poses enormous challenges for evaluation. We perform the first targeted study of the evaluation of long-form answers, covering both human and automatic evaluation practices. We hire domain experts in seven areas to provide preference judgments over pairs of answers, along with free-form justifications for their choices. We present a careful analysis of experts' evaluation, which focuses on new aspects such as the comprehensiveness of the answer. Next, we examine automatic text generation metric"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.18201","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/2305.18201/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":"2305.18201","created_at":"2026-07-05T06:14:53.389803+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.18201v1","created_at":"2026-07-05T06:14:53.389803+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.18201","created_at":"2026-07-05T06:14:53.389803+00:00"},{"alias_kind":"pith_short_12","alias_value":"G4TL5LADVM5E","created_at":"2026-07-05T06:14:53.389803+00:00"},{"alias_kind":"pith_short_16","alias_value":"G4TL5LADVM5EJCPR","created_at":"2026-07-05T06:14:53.389803+00:00"},{"alias_kind":"pith_short_8","alias_value":"G4TL5LAD","created_at":"2026-07-05T06:14:53.389803+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05694","citing_title":"Beyond Heuristic Tuning: Power-Calibrated LLM Watermarking","ref_index":66,"is_internal_anchor":true},{"citing_arxiv_id":"2511.13131","citing_title":"MM-Telco: Benchmarks and Multimodal Large Language Models for Telecom Applications","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2310.11511","citing_title":"Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection","ref_index":157,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G4TL5LADVM5EJCPRJCYFRSL6I4","json":"https://pith.science/pith/G4TL5LADVM5EJCPRJCYFRSL6I4.json","graph_json":"https://pith.science/api/pith-number/G4TL5LADVM5EJCPRJCYFRSL6I4/graph.json","events_json":"https://pith.science/api/pith-number/G4TL5LADVM5EJCPRJCYFRSL6I4/events.json","paper":"https://pith.science/paper/G4TL5LAD"},"agent_actions":{"view_html":"https://pith.science/pith/G4TL5LADVM5EJCPRJCYFRSL6I4","download_json":"https://pith.science/pith/G4TL5LADVM5EJCPRJCYFRSL6I4.json","view_paper":"https://pith.science/paper/G4TL5LAD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.18201&json=true","fetch_graph":"https://pith.science/api/pith-number/G4TL5LADVM5EJCPRJCYFRSL6I4/graph.json","fetch_events":"https://pith.science/api/pith-number/G4TL5LADVM5EJCPRJCYFRSL6I4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G4TL5LADVM5EJCPRJCYFRSL6I4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G4TL5LADVM5EJCPRJCYFRSL6I4/action/storage_attestation","attest_author":"https://pith.science/pith/G4TL5LADVM5EJCPRJCYFRSL6I4/action/author_attestation","sign_citation":"https://pith.science/pith/G4TL5LADVM5EJCPRJCYFRSL6I4/action/citation_signature","submit_replication":"https://pith.science/pith/G4TL5LADVM5EJCPRJCYFRSL6I4/action/replication_record"}},"created_at":"2026-07-05T06:14:53.389803+00:00","updated_at":"2026-07-05T06:14:53.389803+00:00"}