{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:GWRLFIWRUDKIHHFRT5N2C6NY54","short_pith_number":"pith:GWRLFIWR","schema_version":"1.0","canonical_sha256":"35a2b2a2d1a0d4839cb19f5ba179b8ef06b9badf7342438cbbba841ad7692d5c","source":{"kind":"arxiv","id":"2305.03117","version":2},"attestation_state":"computed","paper":{"title":"Are Human Explanations Always Helpful? Towards Objective Evaluation of Human Natural Language Explanations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bingsheng Yao, Dakuo Wang, James Hendler, Lucian Popa, Prithviraj Sen","submitted_at":"2023-05-04T19:31:50Z","abstract_excerpt":"Human-annotated labels and explanations are critical for training explainable NLP models. However, unlike human-annotated labels whose quality is easier to calibrate (e.g., with a majority vote), human-crafted free-form explanations can be quite subjective. Before blindly using them as ground truth to train ML models, a vital question needs to be asked: How do we evaluate a human-annotated explanation's quality? In this paper, we build on the view that the quality of a human-annotated explanation can be measured based on its helpfulness (or impairment) to the ML models' performance for the des"},"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":"2305.03117","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-04T19:31:50Z","cross_cats_sorted":[],"title_canon_sha256":"6cbcd713f6c81c368c26644b73bbad7d9f31874eb637b5ffcb81abc38d41e4eb","abstract_canon_sha256":"427239ce76107967a9b989ab83102881c8945384f7e46bfdba98f29e450cfcd8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:12:02.375887Z","signature_b64":"v9TtfobpUlGPg/rwWjmfoKmT65kD3gMLy3VrErRlANGyRy7rUXzV1+WgbWXEWan9DVkBf4O5B3cOHM/OaI1bDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"35a2b2a2d1a0d4839cb19f5ba179b8ef06b9badf7342438cbbba841ad7692d5c","last_reissued_at":"2026-07-05T06:12:02.375518Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:12:02.375518Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Are Human Explanations Always Helpful? Towards Objective Evaluation of Human Natural Language Explanations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bingsheng Yao, Dakuo Wang, James Hendler, Lucian Popa, Prithviraj Sen","submitted_at":"2023-05-04T19:31:50Z","abstract_excerpt":"Human-annotated labels and explanations are critical for training explainable NLP models. However, unlike human-annotated labels whose quality is easier to calibrate (e.g., with a majority vote), human-crafted free-form explanations can be quite subjective. Before blindly using them as ground truth to train ML models, a vital question needs to be asked: How do we evaluate a human-annotated explanation's quality? In this paper, we build on the view that the quality of a human-annotated explanation can be measured based on its helpfulness (or impairment) to the ML models' performance for the des"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.03117","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/2305.03117/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":"2305.03117","created_at":"2026-07-05T06:12:02.375580+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.03117v2","created_at":"2026-07-05T06:12:02.375580+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.03117","created_at":"2026-07-05T06:12:02.375580+00:00"},{"alias_kind":"pith_short_12","alias_value":"GWRLFIWRUDKI","created_at":"2026-07-05T06:12:02.375580+00:00"},{"alias_kind":"pith_short_16","alias_value":"GWRLFIWRUDKIHHFR","created_at":"2026-07-05T06:12:02.375580+00:00"},{"alias_kind":"pith_short_8","alias_value":"GWRLFIWR","created_at":"2026-07-05T06:12:02.375580+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.24147","citing_title":"Rationales Are Not Silver Bullets: Measuring the Impact of Rationales on Model Performance and Reliability","ref_index":56,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GWRLFIWRUDKIHHFRT5N2C6NY54","json":"https://pith.science/pith/GWRLFIWRUDKIHHFRT5N2C6NY54.json","graph_json":"https://pith.science/api/pith-number/GWRLFIWRUDKIHHFRT5N2C6NY54/graph.json","events_json":"https://pith.science/api/pith-number/GWRLFIWRUDKIHHFRT5N2C6NY54/events.json","paper":"https://pith.science/paper/GWRLFIWR"},"agent_actions":{"view_html":"https://pith.science/pith/GWRLFIWRUDKIHHFRT5N2C6NY54","download_json":"https://pith.science/pith/GWRLFIWRUDKIHHFRT5N2C6NY54.json","view_paper":"https://pith.science/paper/GWRLFIWR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.03117&json=true","fetch_graph":"https://pith.science/api/pith-number/GWRLFIWRUDKIHHFRT5N2C6NY54/graph.json","fetch_events":"https://pith.science/api/pith-number/GWRLFIWRUDKIHHFRT5N2C6NY54/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GWRLFIWRUDKIHHFRT5N2C6NY54/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GWRLFIWRUDKIHHFRT5N2C6NY54/action/storage_attestation","attest_author":"https://pith.science/pith/GWRLFIWRUDKIHHFRT5N2C6NY54/action/author_attestation","sign_citation":"https://pith.science/pith/GWRLFIWRUDKIHHFRT5N2C6NY54/action/citation_signature","submit_replication":"https://pith.science/pith/GWRLFIWRUDKIHHFRT5N2C6NY54/action/replication_record"}},"created_at":"2026-07-05T06:12:02.375580+00:00","updated_at":"2026-07-05T06:12:02.375580+00:00"}