{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:5ZU33PGJSA7A2DL63WHSIOS4NX","short_pith_number":"pith:5ZU33PGJ","schema_version":"1.0","canonical_sha256":"ee69bdbcc9903e0d0d7edd8f243a5c6dc1908002602be557d87436926b87d3aa","source":{"kind":"arxiv","id":"1905.05778","version":3},"attestation_state":"computed","paper":{"title":"Misleading Failures of Partial-input Baselines","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Eric Wallace, Jordan Boyd-Graber, Shi Feng","submitted_at":"2019-05-14T18:01:41Z","abstract_excerpt":"Recent work establishes dataset difficulty and removes annotation artifacts via partial-input baselines (e.g., hypothesis-only models for SNLI or question-only models for VQA). When a partial-input baseline gets high accuracy, a dataset is cheatable. However, the converse is not necessarily true: the failure of a partial-input baseline does not mean a dataset is free of artifacts. To illustrate this, we first design artificial datasets which contain trivial patterns in the full input that are undetectable by any partial-input model. Next, we identify such artifacts in the SNLI dataset - a hypo"},"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":"1905.05778","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-05-14T18:01:41Z","cross_cats_sorted":["cs.AI","cs.CL","stat.ML"],"title_canon_sha256":"59fec97401a47c02bf4f8a3df51cf77a214811db0501735aae70fcbb8b66aa9a","abstract_canon_sha256":"e59605c4cd99d3d0cd18acad46f7a822978c362a9bd5129a17112e0c0ccb41de"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:43:02.099344Z","signature_b64":"tuOj6ZfENE8Q+QmRHxjRWdFUcXJ78t290j6FsX7nwoqg7kdsjjdX5PEm9B5xoCaopgMpadNWIVfQmBtGcs+MCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ee69bdbcc9903e0d0d7edd8f243a5c6dc1908002602be557d87436926b87d3aa","last_reissued_at":"2026-05-17T23:43:02.098891Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:43:02.098891Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Misleading Failures of Partial-input Baselines","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Eric Wallace, Jordan Boyd-Graber, Shi Feng","submitted_at":"2019-05-14T18:01:41Z","abstract_excerpt":"Recent work establishes dataset difficulty and removes annotation artifacts via partial-input baselines (e.g., hypothesis-only models for SNLI or question-only models for VQA). When a partial-input baseline gets high accuracy, a dataset is cheatable. However, the converse is not necessarily true: the failure of a partial-input baseline does not mean a dataset is free of artifacts. To illustrate this, we first design artificial datasets which contain trivial patterns in the full input that are undetectable by any partial-input model. Next, we identify such artifacts in the SNLI dataset - a hypo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1905.05778","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"1905.05778","created_at":"2026-05-17T23:43:02.098965+00:00"},{"alias_kind":"arxiv_version","alias_value":"1905.05778v3","created_at":"2026-05-17T23:43:02.098965+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1905.05778","created_at":"2026-05-17T23:43:02.098965+00:00"},{"alias_kind":"pith_short_12","alias_value":"5ZU33PGJSA7A","created_at":"2026-05-18T12:33:10.108867+00:00"},{"alias_kind":"pith_short_16","alias_value":"5ZU33PGJSA7A2DL6","created_at":"2026-05-18T12:33:10.108867+00:00"},{"alias_kind":"pith_short_8","alias_value":"5ZU33PGJ","created_at":"2026-05-18T12:33:10.108867+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/5ZU33PGJSA7A2DL63WHSIOS4NX","json":"https://pith.science/pith/5ZU33PGJSA7A2DL63WHSIOS4NX.json","graph_json":"https://pith.science/api/pith-number/5ZU33PGJSA7A2DL63WHSIOS4NX/graph.json","events_json":"https://pith.science/api/pith-number/5ZU33PGJSA7A2DL63WHSIOS4NX/events.json","paper":"https://pith.science/paper/5ZU33PGJ"},"agent_actions":{"view_html":"https://pith.science/pith/5ZU33PGJSA7A2DL63WHSIOS4NX","download_json":"https://pith.science/pith/5ZU33PGJSA7A2DL63WHSIOS4NX.json","view_paper":"https://pith.science/paper/5ZU33PGJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1905.05778&json=true","fetch_graph":"https://pith.science/api/pith-number/5ZU33PGJSA7A2DL63WHSIOS4NX/graph.json","fetch_events":"https://pith.science/api/pith-number/5ZU33PGJSA7A2DL63WHSIOS4NX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5ZU33PGJSA7A2DL63WHSIOS4NX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5ZU33PGJSA7A2DL63WHSIOS4NX/action/storage_attestation","attest_author":"https://pith.science/pith/5ZU33PGJSA7A2DL63WHSIOS4NX/action/author_attestation","sign_citation":"https://pith.science/pith/5ZU33PGJSA7A2DL63WHSIOS4NX/action/citation_signature","submit_replication":"https://pith.science/pith/5ZU33PGJSA7A2DL63WHSIOS4NX/action/replication_record"}},"created_at":"2026-05-17T23:43:02.098965+00:00","updated_at":"2026-05-17T23:43:02.098965+00:00"}