{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:LD3LIH44WCA7ZHOD56DQYYV36O","short_pith_number":"pith:LD3LIH44","canonical_record":{"source":{"id":"2202.08048","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-02-16T13:23:14Z","cross_cats_sorted":["cs.AI","cs.IT","math.IT"],"title_canon_sha256":"eb399875b96e5e0b6c998a3c588f366d2e7db6eaa50dbc86a6aa39475a2795cc","abstract_canon_sha256":"30742052fcbbfc9f13a8a9e8c96e0e2000e7ea3f46e60608e8809b6e0e01e384"},"schema_version":"1.0"},"canonical_sha256":"58f6b41f9cb081fc9dc3ef870c62bbf3ad35e96f0c0921d65f22147b6ae14dcf","source":{"kind":"arxiv","id":"2202.08048","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2202.08048","created_at":"2026-07-05T04:56:34Z"},{"alias_kind":"arxiv_version","alias_value":"2202.08048v2","created_at":"2026-07-05T04:56:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.08048","created_at":"2026-07-05T04:56:34Z"},{"alias_kind":"pith_short_12","alias_value":"LD3LIH44WCA7","created_at":"2026-07-05T04:56:34Z"},{"alias_kind":"pith_short_16","alias_value":"LD3LIH44WCA7ZHOD","created_at":"2026-07-05T04:56:34Z"},{"alias_kind":"pith_short_8","alias_value":"LD3LIH44","created_at":"2026-07-05T04:56:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:LD3LIH44WCA7ZHOD56DQYYV36O","target":"record","payload":{"canonical_record":{"source":{"id":"2202.08048","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-02-16T13:23:14Z","cross_cats_sorted":["cs.AI","cs.IT","math.IT"],"title_canon_sha256":"eb399875b96e5e0b6c998a3c588f366d2e7db6eaa50dbc86a6aa39475a2795cc","abstract_canon_sha256":"30742052fcbbfc9f13a8a9e8c96e0e2000e7ea3f46e60608e8809b6e0e01e384"},"schema_version":"1.0"},"canonical_sha256":"58f6b41f9cb081fc9dc3ef870c62bbf3ad35e96f0c0921d65f22147b6ae14dcf","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:56:34.915677Z","signature_b64":"LqJfbT5FFSF6cjYHxeyYNcvD6+pDz7jXNyrqx70C8dR4HSevD3oeQdLQ81RGysKxNasNr47HAIqC2wBPTNgqCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"58f6b41f9cb081fc9dc3ef870c62bbf3ad35e96f0c0921d65f22147b6ae14dcf","last_reissued_at":"2026-07-05T04:56:34.915229Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:56:34.915229Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2202.08048","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:56:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KohYEUCtynFpRCXFInwek+qEd+oFS91BFkkopB1B2UQTL+fGr+3ZgPQUO0jQvg7zeBeHHlukdezc7y46R8juDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T21:07:23.403653Z"},"content_sha256":"c481e3247d73b9556c9a98d280c0ced6027d3a8c67b0f3ba876e122269642f7e","schema_version":"1.0","event_id":"sha256:c481e3247d73b9556c9a98d280c0ced6027d3a8c67b0f3ba876e122269642f7e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:LD3LIH44WCA7ZHOD56DQYYV36O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Decorrelate Irrelevant, Purify Relevant: Overcome Textual Spurious Correlations from a Feature Perspective","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.IT","math.IT"],"primary_cat":"cs.CL","authors_text":"Junjie Shan, Qi Zhang, Rui Zheng, Shihan Dou, Songyang Gao, Ting Wu, Xuanjing Huang, Yueming Wu","submitted_at":"2022-02-16T13:23:14Z","abstract_excerpt":"Natural language understanding (NLU) models tend to rely on spurious correlations (i.e., dataset bias) to achieve high performance on in-distribution datasets but poor performance on out-of-distribution ones. Most of the existing debiasing methods often identify and weaken these samples with biased features (i.e., superficial surface features that cause such spurious correlations). However, down-weighting these samples obstructs the model in learning from the non-biased parts of these samples. To tackle this challenge, in this paper, we propose to eliminate spurious correlations in a fine-grai"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.08048","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/2202.08048/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:56:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KxRv3PXs22b6WWrCKynJmvrYlwM+bpiIVFCAtCN3u6qbtfYElgjB+lRcWC521ZXQURzcnerf7YiCDCqYMdHABQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T21:07:23.404333Z"},"content_sha256":"a1ab06277bfc30f4b13d8c94d97e233558935ee707cf34d98342ce833c5aef61","schema_version":"1.0","event_id":"sha256:a1ab06277bfc30f4b13d8c94d97e233558935ee707cf34d98342ce833c5aef61"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LD3LIH44WCA7ZHOD56DQYYV36O/bundle.json","state_url":"https://pith.science/pith/LD3LIH44WCA7ZHOD56DQYYV36O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LD3LIH44WCA7ZHOD56DQYYV36O/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-06T21:07:23Z","links":{"resolver":"https://pith.science/pith/LD3LIH44WCA7ZHOD56DQYYV36O","bundle":"https://pith.science/pith/LD3LIH44WCA7ZHOD56DQYYV36O/bundle.json","state":"https://pith.science/pith/LD3LIH44WCA7ZHOD56DQYYV36O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LD3LIH44WCA7ZHOD56DQYYV36O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:LD3LIH44WCA7ZHOD56DQYYV36O","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"30742052fcbbfc9f13a8a9e8c96e0e2000e7ea3f46e60608e8809b6e0e01e384","cross_cats_sorted":["cs.AI","cs.IT","math.IT"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-02-16T13:23:14Z","title_canon_sha256":"eb399875b96e5e0b6c998a3c588f366d2e7db6eaa50dbc86a6aa39475a2795cc"},"schema_version":"1.0","source":{"id":"2202.08048","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2202.08048","created_at":"2026-07-05T04:56:34Z"},{"alias_kind":"arxiv_version","alias_value":"2202.08048v2","created_at":"2026-07-05T04:56:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.08048","created_at":"2026-07-05T04:56:34Z"},{"alias_kind":"pith_short_12","alias_value":"LD3LIH44WCA7","created_at":"2026-07-05T04:56:34Z"},{"alias_kind":"pith_short_16","alias_value":"LD3LIH44WCA7ZHOD","created_at":"2026-07-05T04:56:34Z"},{"alias_kind":"pith_short_8","alias_value":"LD3LIH44","created_at":"2026-07-05T04:56:34Z"}],"graph_snapshots":[{"event_id":"sha256:a1ab06277bfc30f4b13d8c94d97e233558935ee707cf34d98342ce833c5aef61","target":"graph","created_at":"2026-07-05T04:56:34Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2202.08048/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Natural language understanding (NLU) models tend to rely on spurious correlations (i.e., dataset bias) to achieve high performance on in-distribution datasets but poor performance on out-of-distribution ones. Most of the existing debiasing methods often identify and weaken these samples with biased features (i.e., superficial surface features that cause such spurious correlations). However, down-weighting these samples obstructs the model in learning from the non-biased parts of these samples. To tackle this challenge, in this paper, we propose to eliminate spurious correlations in a fine-grai","authors_text":"Junjie Shan, Qi Zhang, Rui Zheng, Shihan Dou, Songyang Gao, Ting Wu, Xuanjing Huang, Yueming Wu","cross_cats":["cs.AI","cs.IT","math.IT"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-02-16T13:23:14Z","title":"Decorrelate Irrelevant, Purify Relevant: Overcome Textual Spurious Correlations from a Feature Perspective"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.08048","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:c481e3247d73b9556c9a98d280c0ced6027d3a8c67b0f3ba876e122269642f7e","target":"record","created_at":"2026-07-05T04:56:34Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"30742052fcbbfc9f13a8a9e8c96e0e2000e7ea3f46e60608e8809b6e0e01e384","cross_cats_sorted":["cs.AI","cs.IT","math.IT"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-02-16T13:23:14Z","title_canon_sha256":"eb399875b96e5e0b6c998a3c588f366d2e7db6eaa50dbc86a6aa39475a2795cc"},"schema_version":"1.0","source":{"id":"2202.08048","kind":"arxiv","version":2}},"canonical_sha256":"58f6b41f9cb081fc9dc3ef870c62bbf3ad35e96f0c0921d65f22147b6ae14dcf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"58f6b41f9cb081fc9dc3ef870c62bbf3ad35e96f0c0921d65f22147b6ae14dcf","first_computed_at":"2026-07-05T04:56:34.915229Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:56:34.915229Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LqJfbT5FFSF6cjYHxeyYNcvD6+pDz7jXNyrqx70C8dR4HSevD3oeQdLQ81RGysKxNasNr47HAIqC2wBPTNgqCw==","signature_status":"signed_v1","signed_at":"2026-07-05T04:56:34.915677Z","signed_message":"canonical_sha256_bytes"},"source_id":"2202.08048","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c481e3247d73b9556c9a98d280c0ced6027d3a8c67b0f3ba876e122269642f7e","sha256:a1ab06277bfc30f4b13d8c94d97e233558935ee707cf34d98342ce833c5aef61"],"state_sha256":"a463c706c502e3190b15e5c139cd55263b92b20d6ca73305b469d2f528d2b884"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6QZqYeuV/9m3IGdkwemyX57lQE231XuCuQtkSlREMyMJLUB0gTtk4eCTxjv3CKTSrIBJ5ub+RH5qvJM6IfxAAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T21:07:23.409322Z","bundle_sha256":"6c20a67dd536df09fa74a81c7f4c16ea1e0704576d2d85f760f364cb00a13b34"}}