{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:QQKPVU4FHEXD3NFWN4ZUPNHBXB","short_pith_number":"pith:QQKPVU4F","canonical_record":{"source":{"id":"2306.04949","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-08T05:44:06Z","cross_cats_sorted":[],"title_canon_sha256":"f052f2797d7e50dc2803b3f77b9d1f574e8aa03c1510c051b11d2bdc592a9287","abstract_canon_sha256":"12b5c754f937cf4454831355f2fcb036dc416dff2f02b6a2f973dac3729f2139"},"schema_version":"1.0"},"canonical_sha256":"8414fad385392e3db4b66f3347b4e1b8532b3cc0b73a15e2cd96765aa80a0f5b","source":{"kind":"arxiv","id":"2306.04949","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.04949","created_at":"2026-07-05T07:06:29Z"},{"alias_kind":"arxiv_version","alias_value":"2306.04949v2","created_at":"2026-07-05T07:06:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.04949","created_at":"2026-07-05T07:06:29Z"},{"alias_kind":"pith_short_12","alias_value":"QQKPVU4FHEXD","created_at":"2026-07-05T07:06:29Z"},{"alias_kind":"pith_short_16","alias_value":"QQKPVU4FHEXD3NFW","created_at":"2026-07-05T07:06:29Z"},{"alias_kind":"pith_short_8","alias_value":"QQKPVU4F","created_at":"2026-07-05T07:06:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:QQKPVU4FHEXD3NFWN4ZUPNHBXB","target":"record","payload":{"canonical_record":{"source":{"id":"2306.04949","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-08T05:44:06Z","cross_cats_sorted":[],"title_canon_sha256":"f052f2797d7e50dc2803b3f77b9d1f574e8aa03c1510c051b11d2bdc592a9287","abstract_canon_sha256":"12b5c754f937cf4454831355f2fcb036dc416dff2f02b6a2f973dac3729f2139"},"schema_version":"1.0"},"canonical_sha256":"8414fad385392e3db4b66f3347b4e1b8532b3cc0b73a15e2cd96765aa80a0f5b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:06:29.617073Z","signature_b64":"EEDtgru1zcc2FQWBbXmOAlvP7W4k+ZyIuhHhntBwAo2HvzwybzjKug2gEmlXsFouDO8MAwmgjxkd9Bm9vNHZAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8414fad385392e3db4b66f3347b4e1b8532b3cc0b73a15e2cd96765aa80a0f5b","last_reissued_at":"2026-07-05T07:06:29.616586Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:06:29.616586Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2306.04949","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":"2026-07-05T07:06:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SY0jihnPVfZJJc5VvhWGoo8fy244N/NH2guepbx9O/bD35HQoQtrnQegXBhLgZggvgDRmzBeKWMuaRIDY+TyAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T20:59:09.178920Z"},"content_sha256":"42cad42f10e09f1939c718444c2a0a143524c7320df68bf95bc7e98026b39019","schema_version":"1.0","event_id":"sha256:42cad42f10e09f1939c718444c2a0a143524c7320df68bf95bc7e98026b39019"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:QQKPVU4FHEXD3NFWN4ZUPNHBXB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Robust Learning with Progressive Data Expansion Against Spurious Correlation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Baharan Mirzasoleiman, Quanquan Gu, Yihe Deng, Yu Yang","submitted_at":"2023-06-08T05:44:06Z","abstract_excerpt":"While deep learning models have shown remarkable performance in various tasks, they are susceptible to learning non-generalizable spurious features rather than the core features that are genuinely correlated to the true label. In this paper, beyond existing analyses of linear models, we theoretically examine the learning process of a two-layer nonlinear convolutional neural network in the presence of spurious features. Our analysis suggests that imbalanced data groups and easily learnable spurious features can lead to the dominance of spurious features during the learning process. In light of "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.04949","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/2306.04949/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":"2026-07-05T07:06:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"q6SsVHAzS703q88V0jnabzxrEM0YCuy79LyjFCO53mZEPbZasKiBRAJEPkNauUvOLIQFbNqVMX5rWsUeqofSAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T20:59:09.179450Z"},"content_sha256":"636dbc3e569e368e0191fbc15e33b67bfd8df5d088693ba695589850cae298cd","schema_version":"1.0","event_id":"sha256:636dbc3e569e368e0191fbc15e33b67bfd8df5d088693ba695589850cae298cd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QQKPVU4FHEXD3NFWN4ZUPNHBXB/bundle.json","state_url":"https://pith.science/pith/QQKPVU4FHEXD3NFWN4ZUPNHBXB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QQKPVU4FHEXD3NFWN4ZUPNHBXB/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":"2026-08-07T20:59:09Z","links":{"resolver":"https://pith.science/pith/QQKPVU4FHEXD3NFWN4ZUPNHBXB","bundle":"https://pith.science/pith/QQKPVU4FHEXD3NFWN4ZUPNHBXB/bundle.json","state":"https://pith.science/pith/QQKPVU4FHEXD3NFWN4ZUPNHBXB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QQKPVU4FHEXD3NFWN4ZUPNHBXB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:QQKPVU4FHEXD3NFWN4ZUPNHBXB","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":"12b5c754f937cf4454831355f2fcb036dc416dff2f02b6a2f973dac3729f2139","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-08T05:44:06Z","title_canon_sha256":"f052f2797d7e50dc2803b3f77b9d1f574e8aa03c1510c051b11d2bdc592a9287"},"schema_version":"1.0","source":{"id":"2306.04949","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.04949","created_at":"2026-07-05T07:06:29Z"},{"alias_kind":"arxiv_version","alias_value":"2306.04949v2","created_at":"2026-07-05T07:06:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.04949","created_at":"2026-07-05T07:06:29Z"},{"alias_kind":"pith_short_12","alias_value":"QQKPVU4FHEXD","created_at":"2026-07-05T07:06:29Z"},{"alias_kind":"pith_short_16","alias_value":"QQKPVU4FHEXD3NFW","created_at":"2026-07-05T07:06:29Z"},{"alias_kind":"pith_short_8","alias_value":"QQKPVU4F","created_at":"2026-07-05T07:06:29Z"}],"graph_snapshots":[{"event_id":"sha256:636dbc3e569e368e0191fbc15e33b67bfd8df5d088693ba695589850cae298cd","target":"graph","created_at":"2026-07-05T07:06:29Z","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/2306.04949/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While deep learning models have shown remarkable performance in various tasks, they are susceptible to learning non-generalizable spurious features rather than the core features that are genuinely correlated to the true label. In this paper, beyond existing analyses of linear models, we theoretically examine the learning process of a two-layer nonlinear convolutional neural network in the presence of spurious features. Our analysis suggests that imbalanced data groups and easily learnable spurious features can lead to the dominance of spurious features during the learning process. In light of ","authors_text":"Baharan Mirzasoleiman, Quanquan Gu, Yihe Deng, Yu Yang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-08T05:44:06Z","title":"Robust Learning with Progressive Data Expansion Against Spurious Correlation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.04949","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":"sha256:42cad42f10e09f1939c718444c2a0a143524c7320df68bf95bc7e98026b39019","target":"record","created_at":"2026-07-05T07:06:29Z","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":"12b5c754f937cf4454831355f2fcb036dc416dff2f02b6a2f973dac3729f2139","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-08T05:44:06Z","title_canon_sha256":"f052f2797d7e50dc2803b3f77b9d1f574e8aa03c1510c051b11d2bdc592a9287"},"schema_version":"1.0","source":{"id":"2306.04949","kind":"arxiv","version":2}},"canonical_sha256":"8414fad385392e3db4b66f3347b4e1b8532b3cc0b73a15e2cd96765aa80a0f5b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8414fad385392e3db4b66f3347b4e1b8532b3cc0b73a15e2cd96765aa80a0f5b","first_computed_at":"2026-07-05T07:06:29.616586Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:06:29.616586Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"EEDtgru1zcc2FQWBbXmOAlvP7W4k+ZyIuhHhntBwAo2HvzwybzjKug2gEmlXsFouDO8MAwmgjxkd9Bm9vNHZAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:06:29.617073Z","signed_message":"canonical_sha256_bytes"},"source_id":"2306.04949","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:42cad42f10e09f1939c718444c2a0a143524c7320df68bf95bc7e98026b39019","sha256:636dbc3e569e368e0191fbc15e33b67bfd8df5d088693ba695589850cae298cd"],"state_sha256":"26ceeac87d2d2a76f4e5535a28c8caea5124938393261383eb3a7a89454e480c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YKSm369m7ZIKOxJzRBG+W/jrJO7iNvC+WEStyQycQOIJ8edBXPFlRUTJNmJSCchg1nz//De0RHOmAQqKfjs/CQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T20:59:09.184057Z","bundle_sha256":"fc0d971fcfd383ce3eae2cf312143a5fa442af36cabdf51a2cf1ce956da8b621"}}