{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:7GEYQQ5GU6MFQM5AL2GPWWKPUH","short_pith_number":"pith:7GEYQQ5G","canonical_record":{"source":{"id":"2107.00753","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-07-01T21:46:43Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"62a2003f3696d940f0883b57597d16306671212a4dad2df1ff007a2afd70727c","abstract_canon_sha256":"ce87ad9a8d7aa7b4e35772a74cd0cae16779ac0b708968dd8a210ec2201a3ccc"},"schema_version":"1.0"},"canonical_sha256":"f9898843a6a7985833a05e8cfb594fa1f259e7de844403a14f4cc04b31cbfe7f","source":{"kind":"arxiv","id":"2107.00753","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2107.00753","created_at":"2026-07-05T04:05:32Z"},{"alias_kind":"arxiv_version","alias_value":"2107.00753v3","created_at":"2026-07-05T04:05:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.00753","created_at":"2026-07-05T04:05:32Z"},{"alias_kind":"pith_short_12","alias_value":"7GEYQQ5GU6MF","created_at":"2026-07-05T04:05:32Z"},{"alias_kind":"pith_short_16","alias_value":"7GEYQQ5GU6MFQM5A","created_at":"2026-07-05T04:05:32Z"},{"alias_kind":"pith_short_8","alias_value":"7GEYQQ5G","created_at":"2026-07-05T04:05:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:7GEYQQ5GU6MFQM5AL2GPWWKPUH","target":"record","payload":{"canonical_record":{"source":{"id":"2107.00753","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-07-01T21:46:43Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"62a2003f3696d940f0883b57597d16306671212a4dad2df1ff007a2afd70727c","abstract_canon_sha256":"ce87ad9a8d7aa7b4e35772a74cd0cae16779ac0b708968dd8a210ec2201a3ccc"},"schema_version":"1.0"},"canonical_sha256":"f9898843a6a7985833a05e8cfb594fa1f259e7de844403a14f4cc04b31cbfe7f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:05:32.375766Z","signature_b64":"MqFX1ECZ1R66B01xl/dEqnbNP5NV2Z2u3nERitoTDVUX8dZSUgS8T1ZLB3meDS4sG8tv/esuASUyYnmxL/0vCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f9898843a6a7985833a05e8cfb594fa1f259e7de844403a14f4cc04b31cbfe7f","last_reissued_at":"2026-07-05T04:05:32.375281Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:05:32.375281Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2107.00753","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:05:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UNLYxgRNpYfyWDYuvkxMBk+9NjuvHbspG0FccAjrTvTo8uPkCyK6aQX4pwVDbB4VnfEy76yZRKLlCM2PjkC3AA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T09:40:57.042519Z"},"content_sha256":"fb39a4749656eaebde1af839a2505e041c586f74b3af1a790c387ec18376cf27","schema_version":"1.0","event_id":"sha256:fb39a4749656eaebde1af839a2505e041c586f74b3af1a790c387ec18376cf27"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:7GEYQQ5GU6MFQM5AL2GPWWKPUH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"An Investigation of the (In)effectiveness of Counterfactually Augmented Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"He He, Nitish Joshi","submitted_at":"2021-07-01T21:46:43Z","abstract_excerpt":"While pretrained language models achieve excellent performance on natural language understanding benchmarks, they tend to rely on spurious correlations and generalize poorly to out-of-distribution (OOD) data. Recent work has explored using counterfactually-augmented data (CAD) -- data generated by minimally perturbing examples to flip the ground-truth label -- to identify robust features that are invariant under distribution shift. However, empirical results using CAD for OOD generalization have been mixed. To explain this discrepancy, we draw insights from a linear Gaussian model and demonstr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.00753","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/2107.00753/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":"2026-07-05T04:05:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"N1qY/dVZvSo+zXGktnUNbW7JfdpDEnT+WdriQv9bZfEqrrZnLUu0NKdm2Hxfm0t9cvQMp7J4LwMz2VEqMo5yBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T09:40:57.043420Z"},"content_sha256":"1a60fa5af81d5fcf706a47aae6caa38de0ae1cd83d7ecd95449ea5157f1e7ced","schema_version":"1.0","event_id":"sha256:1a60fa5af81d5fcf706a47aae6caa38de0ae1cd83d7ecd95449ea5157f1e7ced"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7GEYQQ5GU6MFQM5AL2GPWWKPUH/bundle.json","state_url":"https://pith.science/pith/7GEYQQ5GU6MFQM5AL2GPWWKPUH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7GEYQQ5GU6MFQM5AL2GPWWKPUH/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":"2026-08-01T09:40:57Z","links":{"resolver":"https://pith.science/pith/7GEYQQ5GU6MFQM5AL2GPWWKPUH","bundle":"https://pith.science/pith/7GEYQQ5GU6MFQM5AL2GPWWKPUH/bundle.json","state":"https://pith.science/pith/7GEYQQ5GU6MFQM5AL2GPWWKPUH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7GEYQQ5GU6MFQM5AL2GPWWKPUH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:7GEYQQ5GU6MFQM5AL2GPWWKPUH","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":"ce87ad9a8d7aa7b4e35772a74cd0cae16779ac0b708968dd8a210ec2201a3ccc","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-07-01T21:46:43Z","title_canon_sha256":"62a2003f3696d940f0883b57597d16306671212a4dad2df1ff007a2afd70727c"},"schema_version":"1.0","source":{"id":"2107.00753","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2107.00753","created_at":"2026-07-05T04:05:32Z"},{"alias_kind":"arxiv_version","alias_value":"2107.00753v3","created_at":"2026-07-05T04:05:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.00753","created_at":"2026-07-05T04:05:32Z"},{"alias_kind":"pith_short_12","alias_value":"7GEYQQ5GU6MF","created_at":"2026-07-05T04:05:32Z"},{"alias_kind":"pith_short_16","alias_value":"7GEYQQ5GU6MFQM5A","created_at":"2026-07-05T04:05:32Z"},{"alias_kind":"pith_short_8","alias_value":"7GEYQQ5G","created_at":"2026-07-05T04:05:32Z"}],"graph_snapshots":[{"event_id":"sha256:1a60fa5af81d5fcf706a47aae6caa38de0ae1cd83d7ecd95449ea5157f1e7ced","target":"graph","created_at":"2026-07-05T04:05:32Z","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/2107.00753/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While pretrained language models achieve excellent performance on natural language understanding benchmarks, they tend to rely on spurious correlations and generalize poorly to out-of-distribution (OOD) data. Recent work has explored using counterfactually-augmented data (CAD) -- data generated by minimally perturbing examples to flip the ground-truth label -- to identify robust features that are invariant under distribution shift. However, empirical results using CAD for OOD generalization have been mixed. To explain this discrepancy, we draw insights from a linear Gaussian model and demonstr","authors_text":"He He, Nitish Joshi","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-07-01T21:46:43Z","title":"An Investigation of the (In)effectiveness of Counterfactually Augmented Data"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.00753","kind":"arxiv","version":3},"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:fb39a4749656eaebde1af839a2505e041c586f74b3af1a790c387ec18376cf27","target":"record","created_at":"2026-07-05T04:05:32Z","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":"ce87ad9a8d7aa7b4e35772a74cd0cae16779ac0b708968dd8a210ec2201a3ccc","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-07-01T21:46:43Z","title_canon_sha256":"62a2003f3696d940f0883b57597d16306671212a4dad2df1ff007a2afd70727c"},"schema_version":"1.0","source":{"id":"2107.00753","kind":"arxiv","version":3}},"canonical_sha256":"f9898843a6a7985833a05e8cfb594fa1f259e7de844403a14f4cc04b31cbfe7f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f9898843a6a7985833a05e8cfb594fa1f259e7de844403a14f4cc04b31cbfe7f","first_computed_at":"2026-07-05T04:05:32.375281Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:05:32.375281Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MqFX1ECZ1R66B01xl/dEqnbNP5NV2Z2u3nERitoTDVUX8dZSUgS8T1ZLB3meDS4sG8tv/esuASUyYnmxL/0vCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T04:05:32.375766Z","signed_message":"canonical_sha256_bytes"},"source_id":"2107.00753","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fb39a4749656eaebde1af839a2505e041c586f74b3af1a790c387ec18376cf27","sha256:1a60fa5af81d5fcf706a47aae6caa38de0ae1cd83d7ecd95449ea5157f1e7ced"],"state_sha256":"3b11f18b89b2a95e2587e7f4a2004e45ed0057536f2c3d29caea02f68a438eba"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"q0Ua5s0qKYVDahb5YY6DJrkKnSCicRGBWRu6sLTI9CcjuaYPqSXArcIDK0BI6wPE85vRUGmB0oN84VNP3aRyDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-01T09:40:57.052114Z","bundle_sha256":"b9c022b28497021feb90d4e8c17772970aeab48a15781304271bfd5389697f2c"}}