{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:RUJOJJ7FKYKW36YTJOU7JFQCEX","short_pith_number":"pith:RUJOJJ7F","schema_version":"1.0","canonical_sha256":"8d12e4a7e556156dfb134ba9f4960225cf074009d522360f19c54b7fef88cc59","source":{"kind":"arxiv","id":"1912.00582","version":4},"attestation_state":"computed","paper":{"title":"BLiMP: The Benchmark of Linguistic Minimal Pairs for English","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alex Warstadt, Alicia Parrish, Anhad Mohananey, Haokun Liu, Samuel R. Bowman, Sheng-Fu Wang, Wei Peng","submitted_at":"2019-12-02T05:42:41Z","abstract_excerpt":"We introduce The Benchmark of Linguistic Minimal Pairs (shortened to BLiMP), a challenge set for evaluating what language models (LMs) know about major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each containing 1000 minimal pairs isolating specific contrasts in syntax, morphology, or semantics. The data is automatically generated according to expert-crafted grammars, and aggregate human agreement with the labels is 96.4%. We use it to evaluate n-gram, LSTM, and Transformer (GPT-2 and Transformer-XL) LMs. We find that state-of-the-art models identify morphological cont"},"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":"1912.00582","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-12-02T05:42:41Z","cross_cats_sorted":[],"title_canon_sha256":"dab32b42cf42007baf7a1053797e9a7ab4aaa5340ffdb1268ea17fe70a2812aa","abstract_canon_sha256":"02f04e8409d36fde424c6e74dd0ab9f453416eeb1815d707591ea610be50330d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:41:13.477155Z","signature_b64":"nl57K5bRkgenBzeRJaM45eqE31Pb5gMJBQ+ZFVUmmEVQdMujS1jHwq91eDEoXPXoDOCQjU+eK1t9unwFdtGpDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8d12e4a7e556156dfb134ba9f4960225cf074009d522360f19c54b7fef88cc59","last_reissued_at":"2026-07-05T05:41:13.476741Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:41:13.476741Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BLiMP: The Benchmark of Linguistic Minimal Pairs for English","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alex Warstadt, Alicia Parrish, Anhad Mohananey, Haokun Liu, Samuel R. Bowman, Sheng-Fu Wang, Wei Peng","submitted_at":"2019-12-02T05:42:41Z","abstract_excerpt":"We introduce The Benchmark of Linguistic Minimal Pairs (shortened to BLiMP), a challenge set for evaluating what language models (LMs) know about major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each containing 1000 minimal pairs isolating specific contrasts in syntax, morphology, or semantics. The data is automatically generated according to expert-crafted grammars, and aggregate human agreement with the labels is 96.4%. We use it to evaluate n-gram, LSTM, and Transformer (GPT-2 and Transformer-XL) LMs. We find that state-of-the-art models identify morphological cont"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1912.00582","kind":"arxiv","version":4},"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/1912.00582/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":"1912.00582","created_at":"2026-07-05T05:41:13.476798+00:00"},{"alias_kind":"arxiv_version","alias_value":"1912.00582v4","created_at":"2026-07-05T05:41:13.476798+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1912.00582","created_at":"2026-07-05T05:41:13.476798+00:00"},{"alias_kind":"pith_short_12","alias_value":"RUJOJJ7FKYKW","created_at":"2026-07-05T05:41:13.476798+00:00"},{"alias_kind":"pith_short_16","alias_value":"RUJOJJ7FKYKW36YT","created_at":"2026-07-05T05:41:13.476798+00:00"},{"alias_kind":"pith_short_8","alias_value":"RUJOJJ7F","created_at":"2026-07-05T05:41:13.476798+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26050","citing_title":"Natural Ungrokking: Asymmetric Control of Which Rules Survive Pretraining","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01381","citing_title":"A framework for analyzing concept representations in neural models","ref_index":167,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RUJOJJ7FKYKW36YTJOU7JFQCEX","json":"https://pith.science/pith/RUJOJJ7FKYKW36YTJOU7JFQCEX.json","graph_json":"https://pith.science/api/pith-number/RUJOJJ7FKYKW36YTJOU7JFQCEX/graph.json","events_json":"https://pith.science/api/pith-number/RUJOJJ7FKYKW36YTJOU7JFQCEX/events.json","paper":"https://pith.science/paper/RUJOJJ7F"},"agent_actions":{"view_html":"https://pith.science/pith/RUJOJJ7FKYKW36YTJOU7JFQCEX","download_json":"https://pith.science/pith/RUJOJJ7FKYKW36YTJOU7JFQCEX.json","view_paper":"https://pith.science/paper/RUJOJJ7F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1912.00582&json=true","fetch_graph":"https://pith.science/api/pith-number/RUJOJJ7FKYKW36YTJOU7JFQCEX/graph.json","fetch_events":"https://pith.science/api/pith-number/RUJOJJ7FKYKW36YTJOU7JFQCEX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RUJOJJ7FKYKW36YTJOU7JFQCEX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RUJOJJ7FKYKW36YTJOU7JFQCEX/action/storage_attestation","attest_author":"https://pith.science/pith/RUJOJJ7FKYKW36YTJOU7JFQCEX/action/author_attestation","sign_citation":"https://pith.science/pith/RUJOJJ7FKYKW36YTJOU7JFQCEX/action/citation_signature","submit_replication":"https://pith.science/pith/RUJOJJ7FKYKW36YTJOU7JFQCEX/action/replication_record"}},"created_at":"2026-07-05T05:41:13.476798+00:00","updated_at":"2026-07-05T05:41:13.476798+00:00"}