{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:7QEAI6NYVUFLYMSZAINBY6LGHY","short_pith_number":"pith:7QEAI6NY","schema_version":"1.0","canonical_sha256":"fc080479b8ad0abc3259021a1c79663e3cd55c2eb77bda3fcdc5257bbbdbe8d0","source":{"kind":"arxiv","id":"2210.06456","version":2},"attestation_state":"computed","paper":{"title":"Are Sample-Efficient NLP Models More Robust?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Ananya Kumar, Nelson F. Liu, Percy Liang, Robin Jia","submitted_at":"2022-10-12T17:54:59Z","abstract_excerpt":"Recent results in image classification and extractive question answering have observed that pre-trained models trained on less in-distribution data have better out-of-distribution performance. However, it is unclear how broadly these trends hold. We conduct a large empirical study across three tasks, three broadly-applicable modeling interventions (increasing model size, using a different adaptation method, and pre-training on more data), and 14 diverse datasets to investigate the relationship between sample efficiency (amount of data needed to reach a given ID accuracy) and robustness (how mo"},"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":"2210.06456","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-10-12T17:54:59Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a71f3683075accf817e8a8a56532413fcbc42d8ec9393c7e2199f5156447d2c7","abstract_canon_sha256":"c6c2feb4d77d3e5e7474db0754e3172330b9b7c5defd719afafb7056d62a6344"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:15:43.396463Z","signature_b64":"D69pFWmzPU2bCAYRuV5ukqE+1Hu5iWcLgYZd+UBSlygV2bC73HE2Xu95vje2wtDsfRNFPYl8grBnHfwjbVTEAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fc080479b8ad0abc3259021a1c79663e3cd55c2eb77bda3fcdc5257bbbdbe8d0","last_reissued_at":"2026-07-05T06:15:43.395938Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:15:43.395938Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Are Sample-Efficient NLP Models More Robust?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Ananya Kumar, Nelson F. Liu, Percy Liang, Robin Jia","submitted_at":"2022-10-12T17:54:59Z","abstract_excerpt":"Recent results in image classification and extractive question answering have observed that pre-trained models trained on less in-distribution data have better out-of-distribution performance. However, it is unclear how broadly these trends hold. We conduct a large empirical study across three tasks, three broadly-applicable modeling interventions (increasing model size, using a different adaptation method, and pre-training on more data), and 14 diverse datasets to investigate the relationship between sample efficiency (amount of data needed to reach a given ID accuracy) and robustness (how mo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.06456","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/2210.06456/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":"2210.06456","created_at":"2026-07-05T06:15:43.396007+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.06456v2","created_at":"2026-07-05T06:15:43.396007+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.06456","created_at":"2026-07-05T06:15:43.396007+00:00"},{"alias_kind":"pith_short_12","alias_value":"7QEAI6NYVUFL","created_at":"2026-07-05T06:15:43.396007+00:00"},{"alias_kind":"pith_short_16","alias_value":"7QEAI6NYVUFLYMSZ","created_at":"2026-07-05T06:15:43.396007+00:00"},{"alias_kind":"pith_short_8","alias_value":"7QEAI6NY","created_at":"2026-07-05T06:15:43.396007+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.16912","citing_title":"From Data to Knowledge: Evaluating How Efficiently Language Models Learn Facts","ref_index":18,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7QEAI6NYVUFLYMSZAINBY6LGHY","json":"https://pith.science/pith/7QEAI6NYVUFLYMSZAINBY6LGHY.json","graph_json":"https://pith.science/api/pith-number/7QEAI6NYVUFLYMSZAINBY6LGHY/graph.json","events_json":"https://pith.science/api/pith-number/7QEAI6NYVUFLYMSZAINBY6LGHY/events.json","paper":"https://pith.science/paper/7QEAI6NY"},"agent_actions":{"view_html":"https://pith.science/pith/7QEAI6NYVUFLYMSZAINBY6LGHY","download_json":"https://pith.science/pith/7QEAI6NYVUFLYMSZAINBY6LGHY.json","view_paper":"https://pith.science/paper/7QEAI6NY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.06456&json=true","fetch_graph":"https://pith.science/api/pith-number/7QEAI6NYVUFLYMSZAINBY6LGHY/graph.json","fetch_events":"https://pith.science/api/pith-number/7QEAI6NYVUFLYMSZAINBY6LGHY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7QEAI6NYVUFLYMSZAINBY6LGHY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7QEAI6NYVUFLYMSZAINBY6LGHY/action/storage_attestation","attest_author":"https://pith.science/pith/7QEAI6NYVUFLYMSZAINBY6LGHY/action/author_attestation","sign_citation":"https://pith.science/pith/7QEAI6NYVUFLYMSZAINBY6LGHY/action/citation_signature","submit_replication":"https://pith.science/pith/7QEAI6NYVUFLYMSZAINBY6LGHY/action/replication_record"}},"created_at":"2026-07-05T06:15:43.396007+00:00","updated_at":"2026-07-05T06:15:43.396007+00:00"}