{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:FYUUIDVVE3QFNTCWYFTVOW3Q2I","short_pith_number":"pith:FYUUIDVV","canonical_record":{"source":{"id":"2403.19863","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-28T22:17:19Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"1e30dbf9d7c0dbb5fc7019f51a322b1d52c45cab59ab795cafcd67dfc23ab051","abstract_canon_sha256":"ada54c1010a9cb19548016c2eb74c431938d754830f5bffdf92c56693b586365"},"schema_version":"1.0"},"canonical_sha256":"2e29440eb526e056cc56c167575b70d23912c6aefbebc55f2fdfbacdb0560892","source":{"kind":"arxiv","id":"2403.19863","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.19863","created_at":"2026-07-05T09:31:44Z"},{"alias_kind":"arxiv_version","alias_value":"2403.19863v4","created_at":"2026-07-05T09:31:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.19863","created_at":"2026-07-05T09:31:44Z"},{"alias_kind":"pith_short_12","alias_value":"FYUUIDVVE3QF","created_at":"2026-07-05T09:31:44Z"},{"alias_kind":"pith_short_16","alias_value":"FYUUIDVVE3QFNTCW","created_at":"2026-07-05T09:31:44Z"},{"alias_kind":"pith_short_8","alias_value":"FYUUIDVV","created_at":"2026-07-05T09:31:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:FYUUIDVVE3QFNTCWYFTVOW3Q2I","target":"record","payload":{"canonical_record":{"source":{"id":"2403.19863","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-28T22:17:19Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"1e30dbf9d7c0dbb5fc7019f51a322b1d52c45cab59ab795cafcd67dfc23ab051","abstract_canon_sha256":"ada54c1010a9cb19548016c2eb74c431938d754830f5bffdf92c56693b586365"},"schema_version":"1.0"},"canonical_sha256":"2e29440eb526e056cc56c167575b70d23912c6aefbebc55f2fdfbacdb0560892","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:31:44.654642Z","signature_b64":"NGyW+Q11HB6GZpYw9oe/levZvNHNIYexSao3A1HMnoi9UgrSOmTWSwT8JsKErXmM2K3KML1zcYqq2olgyFj8DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2e29440eb526e056cc56c167575b70d23912c6aefbebc55f2fdfbacdb0560892","last_reissued_at":"2026-07-05T09:31:44.654090Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:31:44.654090Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.19863","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:31:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xXFg1/YqfQ4D8G3C0FEdT2hCXYXZVoAg9a64UKhSid/hDMHYdk4Fuve+XSraH6Uv54n0ZGgjwvx2OzQT2PUfBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-09T05:36:48.434093Z"},"content_sha256":"03846d7a3a97621aaeb46af61d711c19b2002330489c2bf87e2101687d15ec96","schema_version":"1.0","event_id":"sha256:03846d7a3a97621aaeb46af61d711c19b2002330489c2bf87e2101687d15ec96"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:FYUUIDVVE3QFNTCWYFTVOW3Q2I","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"DeNetDM: Debiasing by Network Depth Modulation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Abhra Chaudhuri, Adarsh Kappiyath, Anjan Dutta, Silpa Vadakkeeveetil Sreelatha","submitted_at":"2024-03-28T22:17:19Z","abstract_excerpt":"Neural networks trained on biased datasets tend to inadvertently learn spurious correlations, hindering generalization. We formally prove that (1) samples that exhibit spurious correlations lie on a lower rank manifold relative to the ones that do not; and (2) the depth of a network acts as an implicit regularizer on the rank of the attribute subspace that is encoded in its representations. Leveraging these insights, we present DeNetDM, a novel debiasing method that uses network depth modulation as a way of developing robustness to spurious correlations. Using a training paradigm derived from "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.19863","kind":"arxiv","version":4},"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/2403.19863/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":"2026-07-05T09:31:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fbfpuCO1c61fWhvaWhOuWUL+AWUEtm+prf1d4IEMD8J04OX5dLRrGXRZ8tVazD1kjfg0SsBtPWyS5LmfmdhjBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-09T05:36:48.434748Z"},"content_sha256":"1c3b72f1ee7854cca2f84344d41271da807aadd9d822c50b4df976d225b9dd01","schema_version":"1.0","event_id":"sha256:1c3b72f1ee7854cca2f84344d41271da807aadd9d822c50b4df976d225b9dd01"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FYUUIDVVE3QFNTCWYFTVOW3Q2I/bundle.json","state_url":"https://pith.science/pith/FYUUIDVVE3QFNTCWYFTVOW3Q2I/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FYUUIDVVE3QFNTCWYFTVOW3Q2I/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":"2026-07-09T05:36:48Z","links":{"resolver":"https://pith.science/pith/FYUUIDVVE3QFNTCWYFTVOW3Q2I","bundle":"https://pith.science/pith/FYUUIDVVE3QFNTCWYFTVOW3Q2I/bundle.json","state":"https://pith.science/pith/FYUUIDVVE3QFNTCWYFTVOW3Q2I/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FYUUIDVVE3QFNTCWYFTVOW3Q2I/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:FYUUIDVVE3QFNTCWYFTVOW3Q2I","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":"ada54c1010a9cb19548016c2eb74c431938d754830f5bffdf92c56693b586365","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-28T22:17:19Z","title_canon_sha256":"1e30dbf9d7c0dbb5fc7019f51a322b1d52c45cab59ab795cafcd67dfc23ab051"},"schema_version":"1.0","source":{"id":"2403.19863","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.19863","created_at":"2026-07-05T09:31:44Z"},{"alias_kind":"arxiv_version","alias_value":"2403.19863v4","created_at":"2026-07-05T09:31:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.19863","created_at":"2026-07-05T09:31:44Z"},{"alias_kind":"pith_short_12","alias_value":"FYUUIDVVE3QF","created_at":"2026-07-05T09:31:44Z"},{"alias_kind":"pith_short_16","alias_value":"FYUUIDVVE3QFNTCW","created_at":"2026-07-05T09:31:44Z"},{"alias_kind":"pith_short_8","alias_value":"FYUUIDVV","created_at":"2026-07-05T09:31:44Z"}],"graph_snapshots":[{"event_id":"sha256:1c3b72f1ee7854cca2f84344d41271da807aadd9d822c50b4df976d225b9dd01","target":"graph","created_at":"2026-07-05T09:31:44Z","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/2403.19863/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Neural networks trained on biased datasets tend to inadvertently learn spurious correlations, hindering generalization. We formally prove that (1) samples that exhibit spurious correlations lie on a lower rank manifold relative to the ones that do not; and (2) the depth of a network acts as an implicit regularizer on the rank of the attribute subspace that is encoded in its representations. Leveraging these insights, we present DeNetDM, a novel debiasing method that uses network depth modulation as a way of developing robustness to spurious correlations. Using a training paradigm derived from ","authors_text":"Abhra Chaudhuri, Adarsh Kappiyath, Anjan Dutta, Silpa Vadakkeeveetil Sreelatha","cross_cats":["cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-28T22:17:19Z","title":"DeNetDM: Debiasing by Network Depth Modulation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.19863","kind":"arxiv","version":4},"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":"sha256:03846d7a3a97621aaeb46af61d711c19b2002330489c2bf87e2101687d15ec96","target":"record","created_at":"2026-07-05T09:31:44Z","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":"ada54c1010a9cb19548016c2eb74c431938d754830f5bffdf92c56693b586365","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-28T22:17:19Z","title_canon_sha256":"1e30dbf9d7c0dbb5fc7019f51a322b1d52c45cab59ab795cafcd67dfc23ab051"},"schema_version":"1.0","source":{"id":"2403.19863","kind":"arxiv","version":4}},"canonical_sha256":"2e29440eb526e056cc56c167575b70d23912c6aefbebc55f2fdfbacdb0560892","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2e29440eb526e056cc56c167575b70d23912c6aefbebc55f2fdfbacdb0560892","first_computed_at":"2026-07-05T09:31:44.654090Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:31:44.654090Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NGyW+Q11HB6GZpYw9oe/levZvNHNIYexSao3A1HMnoi9UgrSOmTWSwT8JsKErXmM2K3KML1zcYqq2olgyFj8DQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:31:44.654642Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.19863","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:03846d7a3a97621aaeb46af61d711c19b2002330489c2bf87e2101687d15ec96","sha256:1c3b72f1ee7854cca2f84344d41271da807aadd9d822c50b4df976d225b9dd01"],"state_sha256":"f35516b98c02a781b3112fc62c88a67ed950b83da41c833d20cec1816ecd7e9c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3iFkdo3XxE1QBrKbQxyblNMCEgNuCkfm2hJUzvvm7s5fskIlMceV63jIqgYdlCMhYaZzwmADaSy6BbW1ekQNBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-09T05:36:48.438638Z","bundle_sha256":"33780b6ad34d8e3ffb40d95264700ad868ee4d5c6360853d74d23f277722d58b"}}