{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:QRXF5P7F4JIS3TDKRMWGPSSQYD","short_pith_number":"pith:QRXF5P7F","canonical_record":{"source":{"id":"2409.04188","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-09-06T11:05:26Z","cross_cats_sorted":[],"title_canon_sha256":"41cb7cccb6a57031c440fececc871e6a62dc4000b061cd3c3afe6ee00f2e9a6b","abstract_canon_sha256":"bf77f75c14e3498219281126465e9905bfbe3222190a423454942ff8931da2cd"},"schema_version":"1.0"},"canonical_sha256":"846e5ebfe5e2512dcc6a8b2c67ca50c0ea5e20cc17ae20e38f888ee2d9dbe9f7","source":{"kind":"arxiv","id":"2409.04188","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.04188","created_at":"2026-07-05T09:04:02Z"},{"alias_kind":"arxiv_version","alias_value":"2409.04188v1","created_at":"2026-07-05T09:04:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.04188","created_at":"2026-07-05T09:04:02Z"},{"alias_kind":"pith_short_12","alias_value":"QRXF5P7F4JIS","created_at":"2026-07-05T09:04:02Z"},{"alias_kind":"pith_short_16","alias_value":"QRXF5P7F4JIS3TDK","created_at":"2026-07-05T09:04:02Z"},{"alias_kind":"pith_short_8","alias_value":"QRXF5P7F","created_at":"2026-07-05T09:04:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:QRXF5P7F4JIS3TDKRMWGPSSQYD","target":"record","payload":{"canonical_record":{"source":{"id":"2409.04188","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-09-06T11:05:26Z","cross_cats_sorted":[],"title_canon_sha256":"41cb7cccb6a57031c440fececc871e6a62dc4000b061cd3c3afe6ee00f2e9a6b","abstract_canon_sha256":"bf77f75c14e3498219281126465e9905bfbe3222190a423454942ff8931da2cd"},"schema_version":"1.0"},"canonical_sha256":"846e5ebfe5e2512dcc6a8b2c67ca50c0ea5e20cc17ae20e38f888ee2d9dbe9f7","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:04:02.835047Z","signature_b64":"rB2L6GWn5iHA9KXj1xhcdBITqcn+MzQyZiDXwEXkgxlXKArkzCqqVipaej9UjO0bkckY0NC0GKY9flOQKoyACQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"846e5ebfe5e2512dcc6a8b2c67ca50c0ea5e20cc17ae20e38f888ee2d9dbe9f7","last_reissued_at":"2026-07-05T09:04:02.834704Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:04:02.834704Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2409.04188","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:04:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ivyz+Imtbwbprs2uuvRkiAQ+K6rBE5NBnvmpEfhCfjB2rY7M8mPBKPVM3EtEZFn5ypANzCo4lkHD1cg2/CyHAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T20:17:52.262830Z"},"content_sha256":"3ce7f584425dcf0b7de3b852a2821e6b6c39fcd07f336630aea18f04a910ad61","schema_version":"1.0","event_id":"sha256:3ce7f584425dcf0b7de3b852a2821e6b6c39fcd07f336630aea18f04a910ad61"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:QRXF5P7F4JIS3TDKRMWGPSSQYD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Reassessing the Validity of Spurious Correlations Benchmarks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Diane Bouchacourt, Levent Sagun, Samuel J. Bell","submitted_at":"2024-09-06T11:05:26Z","abstract_excerpt":"Neural networks can fail when the data contains spurious correlations. To understand this phenomenon, researchers have proposed numerous spurious correlations benchmarks upon which to evaluate mitigation methods. However, we observe that these benchmarks exhibit substantial disagreement, with the best methods on one benchmark performing poorly on another. We explore this disagreement, and examine benchmark validity by defining three desiderata that a benchmark should satisfy in order to meaningfully evaluate methods. Our results have implications for both benchmarks and mitigations: we find th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.04188","kind":"arxiv","version":1},"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/2409.04188/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":"2026-07-05T09:04:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"13D9qurKBsMUSvxeeVC/07Qe5L/AUUqlQx8VUbYbxWorj3FN/zBBHSltnQpgQ9AlgaakVu8Zbf5tkX9iCBBMAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T20:17:52.263561Z"},"content_sha256":"6cdfdee3ae0db0fbf7079742746f167338d1567e49dcea4a344023d9ef28a782","schema_version":"1.0","event_id":"sha256:6cdfdee3ae0db0fbf7079742746f167338d1567e49dcea4a344023d9ef28a782"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QRXF5P7F4JIS3TDKRMWGPSSQYD/bundle.json","state_url":"https://pith.science/pith/QRXF5P7F4JIS3TDKRMWGPSSQYD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QRXF5P7F4JIS3TDKRMWGPSSQYD/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":"2026-08-10T20:17:52Z","links":{"resolver":"https://pith.science/pith/QRXF5P7F4JIS3TDKRMWGPSSQYD","bundle":"https://pith.science/pith/QRXF5P7F4JIS3TDKRMWGPSSQYD/bundle.json","state":"https://pith.science/pith/QRXF5P7F4JIS3TDKRMWGPSSQYD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QRXF5P7F4JIS3TDKRMWGPSSQYD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:QRXF5P7F4JIS3TDKRMWGPSSQYD","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":"bf77f75c14e3498219281126465e9905bfbe3222190a423454942ff8931da2cd","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-09-06T11:05:26Z","title_canon_sha256":"41cb7cccb6a57031c440fececc871e6a62dc4000b061cd3c3afe6ee00f2e9a6b"},"schema_version":"1.0","source":{"id":"2409.04188","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.04188","created_at":"2026-07-05T09:04:02Z"},{"alias_kind":"arxiv_version","alias_value":"2409.04188v1","created_at":"2026-07-05T09:04:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.04188","created_at":"2026-07-05T09:04:02Z"},{"alias_kind":"pith_short_12","alias_value":"QRXF5P7F4JIS","created_at":"2026-07-05T09:04:02Z"},{"alias_kind":"pith_short_16","alias_value":"QRXF5P7F4JIS3TDK","created_at":"2026-07-05T09:04:02Z"},{"alias_kind":"pith_short_8","alias_value":"QRXF5P7F","created_at":"2026-07-05T09:04:02Z"}],"graph_snapshots":[{"event_id":"sha256:6cdfdee3ae0db0fbf7079742746f167338d1567e49dcea4a344023d9ef28a782","target":"graph","created_at":"2026-07-05T09:04:02Z","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/2409.04188/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Neural networks can fail when the data contains spurious correlations. To understand this phenomenon, researchers have proposed numerous spurious correlations benchmarks upon which to evaluate mitigation methods. However, we observe that these benchmarks exhibit substantial disagreement, with the best methods on one benchmark performing poorly on another. We explore this disagreement, and examine benchmark validity by defining three desiderata that a benchmark should satisfy in order to meaningfully evaluate methods. Our results have implications for both benchmarks and mitigations: we find th","authors_text":"Diane Bouchacourt, Levent Sagun, Samuel J. Bell","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-09-06T11:05:26Z","title":"Reassessing the Validity of Spurious Correlations Benchmarks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.04188","kind":"arxiv","version":1},"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":"sha256:3ce7f584425dcf0b7de3b852a2821e6b6c39fcd07f336630aea18f04a910ad61","target":"record","created_at":"2026-07-05T09:04:02Z","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":"bf77f75c14e3498219281126465e9905bfbe3222190a423454942ff8931da2cd","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-09-06T11:05:26Z","title_canon_sha256":"41cb7cccb6a57031c440fececc871e6a62dc4000b061cd3c3afe6ee00f2e9a6b"},"schema_version":"1.0","source":{"id":"2409.04188","kind":"arxiv","version":1}},"canonical_sha256":"846e5ebfe5e2512dcc6a8b2c67ca50c0ea5e20cc17ae20e38f888ee2d9dbe9f7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"846e5ebfe5e2512dcc6a8b2c67ca50c0ea5e20cc17ae20e38f888ee2d9dbe9f7","first_computed_at":"2026-07-05T09:04:02.834704Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:04:02.834704Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rB2L6GWn5iHA9KXj1xhcdBITqcn+MzQyZiDXwEXkgxlXKArkzCqqVipaej9UjO0bkckY0NC0GKY9flOQKoyACQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:04:02.835047Z","signed_message":"canonical_sha256_bytes"},"source_id":"2409.04188","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3ce7f584425dcf0b7de3b852a2821e6b6c39fcd07f336630aea18f04a910ad61","sha256:6cdfdee3ae0db0fbf7079742746f167338d1567e49dcea4a344023d9ef28a782"],"state_sha256":"8c2a29e4fe518f5489309f6fc8a57a81e31c716143e8e0da7a8817de33a0abc5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eC7Bp9S+HBjNwFiFnfaYQVhq2tz+bZWj85qC2uynt3EQ6IzVAhN9GrIwbpZd+ooDXWeUCL+MxvDJghcT9C1dAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T20:17:52.270855Z","bundle_sha256":"bf68a3bda79938823f964ea3c9b954d4967fc5b75415e68071e820b13847e28c"}}