{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:XQSVHMFPGZWYHYUSNYK7TAX46I","short_pith_number":"pith:XQSVHMFP","schema_version":"1.0","canonical_sha256":"bc2553b0af366d83e2926e15f982fcf21f5edec446fa4e366f9977961c42f712","source":{"kind":"arxiv","id":"2010.03777","version":2},"attestation_state":"computed","paper":{"title":"An Empirical Study on Model-agnostic Debiasing Strategies for Robust Natural Language Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Baobao Chang, Tianyu Liu, Xiaoan Ding, Xin Zheng, Zhifang Sui","submitted_at":"2020-10-08T05:40:45Z","abstract_excerpt":"The prior work on natural language inference (NLI) debiasing mainly targets at one or few known biases while not necessarily making the models more robust. In this paper, we focus on the model-agnostic debiasing strategies and explore how to (or is it possible to) make the NLI models robust to multiple distinct adversarial attacks while keeping or even strengthening the models' generalization power. We firstly benchmark prevailing neural NLI models including pretrained ones on various adversarial datasets. We then try to combat distinct known biases by modifying a mixture of experts (MoE) ense"},"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":"2010.03777","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2020-10-08T05:40:45Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c9a0bde364cd5bc4f65f6747a978c4966b1e0b7618be879aab4b2fab8b03cc0a","abstract_canon_sha256":"6821f248b09235804ac14f6b203cd826b6790009020aa8454d02d5582c3e8361"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:43:42.331525Z","signature_b64":"wgrip4bG5PHj/dh+Tjz3j3DSqRoYg90Mbg4AClCBR28/Uc5MHJET6r1YqRuDnivmfrJXKQNhhomrYzhi8VglDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bc2553b0af366d83e2926e15f982fcf21f5edec446fa4e366f9977961c42f712","last_reissued_at":"2026-07-05T01:43:42.331132Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:43:42.331132Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"An Empirical Study on Model-agnostic Debiasing Strategies for Robust Natural Language Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Baobao Chang, Tianyu Liu, Xiaoan Ding, Xin Zheng, Zhifang Sui","submitted_at":"2020-10-08T05:40:45Z","abstract_excerpt":"The prior work on natural language inference (NLI) debiasing mainly targets at one or few known biases while not necessarily making the models more robust. In this paper, we focus on the model-agnostic debiasing strategies and explore how to (or is it possible to) make the NLI models robust to multiple distinct adversarial attacks while keeping or even strengthening the models' generalization power. We firstly benchmark prevailing neural NLI models including pretrained ones on various adversarial datasets. We then try to combat distinct known biases by modifying a mixture of experts (MoE) ense"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.03777","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/2010.03777/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":"2010.03777","created_at":"2026-07-05T01:43:42.331196+00:00"},{"alias_kind":"arxiv_version","alias_value":"2010.03777v2","created_at":"2026-07-05T01:43:42.331196+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.03777","created_at":"2026-07-05T01:43:42.331196+00:00"},{"alias_kind":"pith_short_12","alias_value":"XQSVHMFPGZWY","created_at":"2026-07-05T01:43:42.331196+00:00"},{"alias_kind":"pith_short_16","alias_value":"XQSVHMFPGZWYHYUS","created_at":"2026-07-05T01:43:42.331196+00:00"},{"alias_kind":"pith_short_8","alias_value":"XQSVHMFP","created_at":"2026-07-05T01:43:42.331196+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/XQSVHMFPGZWYHYUSNYK7TAX46I","json":"https://pith.science/pith/XQSVHMFPGZWYHYUSNYK7TAX46I.json","graph_json":"https://pith.science/api/pith-number/XQSVHMFPGZWYHYUSNYK7TAX46I/graph.json","events_json":"https://pith.science/api/pith-number/XQSVHMFPGZWYHYUSNYK7TAX46I/events.json","paper":"https://pith.science/paper/XQSVHMFP"},"agent_actions":{"view_html":"https://pith.science/pith/XQSVHMFPGZWYHYUSNYK7TAX46I","download_json":"https://pith.science/pith/XQSVHMFPGZWYHYUSNYK7TAX46I.json","view_paper":"https://pith.science/paper/XQSVHMFP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2010.03777&json=true","fetch_graph":"https://pith.science/api/pith-number/XQSVHMFPGZWYHYUSNYK7TAX46I/graph.json","fetch_events":"https://pith.science/api/pith-number/XQSVHMFPGZWYHYUSNYK7TAX46I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XQSVHMFPGZWYHYUSNYK7TAX46I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XQSVHMFPGZWYHYUSNYK7TAX46I/action/storage_attestation","attest_author":"https://pith.science/pith/XQSVHMFPGZWYHYUSNYK7TAX46I/action/author_attestation","sign_citation":"https://pith.science/pith/XQSVHMFPGZWYHYUSNYK7TAX46I/action/citation_signature","submit_replication":"https://pith.science/pith/XQSVHMFPGZWYHYUSNYK7TAX46I/action/replication_record"}},"created_at":"2026-07-05T01:43:42.331196+00:00","updated_at":"2026-07-05T01:43:42.331196+00:00"}