{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:FTDKK4BBMHCB2CEJYZRNVFUMNL","short_pith_number":"pith:FTDKK4BB","schema_version":"1.0","canonical_sha256":"2cc6a5702161c41d0889c662da968c6ac36beadd323eb81c3868872dcecca7a0","source":{"kind":"arxiv","id":"2010.08580","version":3},"attestation_state":"computed","paper":{"title":"Linguistically-Informed Transformations (LIT): A Method for Automatically Generating Contrast Sets","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chuanrong Li, Leo Z. Liu, Lin Shengshuo, Shane Steinert-Threlkeld, Xinyi Wu, Xuhui Zhou","submitted_at":"2020-10-16T18:23:05Z","abstract_excerpt":"Although large-scale pretrained language models, such as BERT and RoBERTa, have achieved superhuman performance on in-distribution test sets, their performance suffers on out-of-distribution test sets (e.g., on contrast sets). Building contrast sets often re-quires human-expert annotation, which is expensive and hard to create on a large scale. In this work, we propose a Linguistically-Informed Transformation (LIT) method to automatically generate contrast sets, which enables practitioners to explore linguistic phenomena of interests as well as compose different phenomena. Experimenting with o"},"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":"2010.08580","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-16T18:23:05Z","cross_cats_sorted":[],"title_canon_sha256":"86ab1f8c8b799d8991158ae9b677ade928e4c2fe0bed2b49ddac27adc099c70f","abstract_canon_sha256":"b728e4f1ce0bdd8c98386667beb9a7bd5c31aa71b2dfa50e0877277fc80a14b2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:51:09.469352Z","signature_b64":"72e8eujv0V2Xkwz4yhxkFk1K5COS86g7Dm3wPps1+Re5vXjFlcaQ/+FxBhLIIda9jg3BX/CaH9pc0ypztzvaDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2cc6a5702161c41d0889c662da968c6ac36beadd323eb81c3868872dcecca7a0","last_reissued_at":"2026-07-05T01:51:09.468690Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:51:09.468690Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Linguistically-Informed Transformations (LIT): A Method for Automatically Generating Contrast Sets","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chuanrong Li, Leo Z. Liu, Lin Shengshuo, Shane Steinert-Threlkeld, Xinyi Wu, Xuhui Zhou","submitted_at":"2020-10-16T18:23:05Z","abstract_excerpt":"Although large-scale pretrained language models, such as BERT and RoBERTa, have achieved superhuman performance on in-distribution test sets, their performance suffers on out-of-distribution test sets (e.g., on contrast sets). Building contrast sets often re-quires human-expert annotation, which is expensive and hard to create on a large scale. In this work, we propose a Linguistically-Informed Transformation (LIT) method to automatically generate contrast sets, which enables practitioners to explore linguistic phenomena of interests as well as compose different phenomena. Experimenting with o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.08580","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2010.08580/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":"2010.08580","created_at":"2026-07-05T01:51:09.468778+00:00"},{"alias_kind":"arxiv_version","alias_value":"2010.08580v3","created_at":"2026-07-05T01:51:09.468778+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.08580","created_at":"2026-07-05T01:51:09.468778+00:00"},{"alias_kind":"pith_short_12","alias_value":"FTDKK4BBMHCB","created_at":"2026-07-05T01:51:09.468778+00:00"},{"alias_kind":"pith_short_16","alias_value":"FTDKK4BBMHCB2CEJ","created_at":"2026-07-05T01:51:09.468778+00:00"},{"alias_kind":"pith_short_8","alias_value":"FTDKK4BB","created_at":"2026-07-05T01:51:09.468778+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2303.09014","citing_title":"ART: Automatic multi-step reasoning and tool-use for large language models","ref_index":78,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FTDKK4BBMHCB2CEJYZRNVFUMNL","json":"https://pith.science/pith/FTDKK4BBMHCB2CEJYZRNVFUMNL.json","graph_json":"https://pith.science/api/pith-number/FTDKK4BBMHCB2CEJYZRNVFUMNL/graph.json","events_json":"https://pith.science/api/pith-number/FTDKK4BBMHCB2CEJYZRNVFUMNL/events.json","paper":"https://pith.science/paper/FTDKK4BB"},"agent_actions":{"view_html":"https://pith.science/pith/FTDKK4BBMHCB2CEJYZRNVFUMNL","download_json":"https://pith.science/pith/FTDKK4BBMHCB2CEJYZRNVFUMNL.json","view_paper":"https://pith.science/paper/FTDKK4BB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2010.08580&json=true","fetch_graph":"https://pith.science/api/pith-number/FTDKK4BBMHCB2CEJYZRNVFUMNL/graph.json","fetch_events":"https://pith.science/api/pith-number/FTDKK4BBMHCB2CEJYZRNVFUMNL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FTDKK4BBMHCB2CEJYZRNVFUMNL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FTDKK4BBMHCB2CEJYZRNVFUMNL/action/storage_attestation","attest_author":"https://pith.science/pith/FTDKK4BBMHCB2CEJYZRNVFUMNL/action/author_attestation","sign_citation":"https://pith.science/pith/FTDKK4BBMHCB2CEJYZRNVFUMNL/action/citation_signature","submit_replication":"https://pith.science/pith/FTDKK4BBMHCB2CEJYZRNVFUMNL/action/replication_record"}},"created_at":"2026-07-05T01:51:09.468778+00:00","updated_at":"2026-07-05T01:51:09.468778+00:00"}