{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:LKR7LU7YBSR3NMJAAKZVBQLK44","short_pith_number":"pith:LKR7LU7Y","canonical_record":{"source":{"id":"2607.03613","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-03T22:03:13Z","cross_cats_sorted":[],"title_canon_sha256":"1772e42cf24d4fe1bb0478613b7027822502ef4f5aea671a8028d3aa5caf8bc4","abstract_canon_sha256":"8e78e89f6d0e1cdb97596da68cec9efa32b5bb32677a8aa4267a68a68998a8eb"},"schema_version":"1.0"},"canonical_sha256":"5aa3f5d3f80ca3b6b12002b350c16ae720f5d2c050dedcee49a49f04468252b4","source":{"kind":"arxiv","id":"2607.03613","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.03613","created_at":"2026-07-07T02:17:57Z"},{"alias_kind":"arxiv_version","alias_value":"2607.03613v1","created_at":"2026-07-07T02:17:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.03613","created_at":"2026-07-07T02:17:57Z"},{"alias_kind":"pith_short_12","alias_value":"LKR7LU7YBSR3","created_at":"2026-07-07T02:17:57Z"},{"alias_kind":"pith_short_16","alias_value":"LKR7LU7YBSR3NMJA","created_at":"2026-07-07T02:17:57Z"},{"alias_kind":"pith_short_8","alias_value":"LKR7LU7Y","created_at":"2026-07-07T02:17:57Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:LKR7LU7YBSR3NMJAAKZVBQLK44","target":"record","payload":{"canonical_record":{"source":{"id":"2607.03613","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-03T22:03:13Z","cross_cats_sorted":[],"title_canon_sha256":"1772e42cf24d4fe1bb0478613b7027822502ef4f5aea671a8028d3aa5caf8bc4","abstract_canon_sha256":"8e78e89f6d0e1cdb97596da68cec9efa32b5bb32677a8aa4267a68a68998a8eb"},"schema_version":"1.0"},"canonical_sha256":"5aa3f5d3f80ca3b6b12002b350c16ae720f5d2c050dedcee49a49f04468252b4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T02:17:57.470798Z","signature_b64":"6ftq6wrNrPKrWxIQXMRT1GRplq/b3RhnyDDgm9Ne0ubLh5AmoLmvi92mh02d1jWnzAxy/X0TuLJETBnDFJDuCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5aa3f5d3f80ca3b6b12002b350c16ae720f5d2c050dedcee49a49f04468252b4","last_reissued_at":"2026-07-07T02:17:57.470032Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T02:17:57.470032Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.03613","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":"2026-07-07T02:17:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OyEYydMQrf5+mI2pPp8Wjm/rIWz59RBFmGwhmSrII7HIkSIViSOwYNCXZ6gN4Whk50IYqvf7BBCEPc9vgBIcCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T23:43:48.773407Z"},"content_sha256":"4691b453af95b596bf6b1a0c1afc3196829286979a3654c2db6946d99ef33568","schema_version":"1.0","event_id":"sha256:4691b453af95b596bf6b1a0c1afc3196829286979a3654c2db6946d99ef33568"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:LKR7LU7YBSR3NMJAAKZVBQLK44","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Implicit Bias of SGD in Multivariate ReLU Networks: Effective Width Collapse","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Guido Mont\\'ufar, Shuang Liang, Tom Jacobs","submitted_at":"2026-07-03T22:03:13Z","abstract_excerpt":"We study the implicit bias of noisy stochastic gradient descent in training wide two-layer ReLU networks for multivariate regression. In a mean-field regime, the training dynamics are approximated by a Wasserstein gradient flow that converges to a unique stationary measure. We characterize the structure of this stationary measure and the predictor it represents. We show that, despite the network being infinitely overparameterized, the learned predictor admits an effectively finite representation: the input weights and biases align along finitely many directions, leading to an effective width c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.03613","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/2607.03613/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":"2026-07-07T02:17:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"L3SxzYKquecMlO0/VStfmSUPEpS3k1cC4hSSAXY+UW6h9LxMalbKSE5vOp6xGQAdiUF8dNfz3+UZAfEbNvogBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T23:43:48.773945Z"},"content_sha256":"01b6b89f8a53e695c11a414d45b0736fa00d17f64c67982be09b823cc2f8a7e5","schema_version":"1.0","event_id":"sha256:01b6b89f8a53e695c11a414d45b0736fa00d17f64c67982be09b823cc2f8a7e5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LKR7LU7YBSR3NMJAAKZVBQLK44/bundle.json","state_url":"https://pith.science/pith/LKR7LU7YBSR3NMJAAKZVBQLK44/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LKR7LU7YBSR3NMJAAKZVBQLK44/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":"2026-08-03T23:43:48Z","links":{"resolver":"https://pith.science/pith/LKR7LU7YBSR3NMJAAKZVBQLK44","bundle":"https://pith.science/pith/LKR7LU7YBSR3NMJAAKZVBQLK44/bundle.json","state":"https://pith.science/pith/LKR7LU7YBSR3NMJAAKZVBQLK44/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LKR7LU7YBSR3NMJAAKZVBQLK44/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:LKR7LU7YBSR3NMJAAKZVBQLK44","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":"8e78e89f6d0e1cdb97596da68cec9efa32b5bb32677a8aa4267a68a68998a8eb","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-03T22:03:13Z","title_canon_sha256":"1772e42cf24d4fe1bb0478613b7027822502ef4f5aea671a8028d3aa5caf8bc4"},"schema_version":"1.0","source":{"id":"2607.03613","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.03613","created_at":"2026-07-07T02:17:57Z"},{"alias_kind":"arxiv_version","alias_value":"2607.03613v1","created_at":"2026-07-07T02:17:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.03613","created_at":"2026-07-07T02:17:57Z"},{"alias_kind":"pith_short_12","alias_value":"LKR7LU7YBSR3","created_at":"2026-07-07T02:17:57Z"},{"alias_kind":"pith_short_16","alias_value":"LKR7LU7YBSR3NMJA","created_at":"2026-07-07T02:17:57Z"},{"alias_kind":"pith_short_8","alias_value":"LKR7LU7Y","created_at":"2026-07-07T02:17:57Z"}],"graph_snapshots":[{"event_id":"sha256:01b6b89f8a53e695c11a414d45b0736fa00d17f64c67982be09b823cc2f8a7e5","target":"graph","created_at":"2026-07-07T02:17:57Z","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/2607.03613/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We study the implicit bias of noisy stochastic gradient descent in training wide two-layer ReLU networks for multivariate regression. In a mean-field regime, the training dynamics are approximated by a Wasserstein gradient flow that converges to a unique stationary measure. We characterize the structure of this stationary measure and the predictor it represents. We show that, despite the network being infinitely overparameterized, the learned predictor admits an effectively finite representation: the input weights and biases align along finitely many directions, leading to an effective width c","authors_text":"Guido Mont\\'ufar, Shuang Liang, Tom Jacobs","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-03T22:03:13Z","title":"Implicit Bias of SGD in Multivariate ReLU Networks: Effective Width Collapse"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.03613","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":"sha256:4691b453af95b596bf6b1a0c1afc3196829286979a3654c2db6946d99ef33568","target":"record","created_at":"2026-07-07T02:17:57Z","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":"8e78e89f6d0e1cdb97596da68cec9efa32b5bb32677a8aa4267a68a68998a8eb","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-03T22:03:13Z","title_canon_sha256":"1772e42cf24d4fe1bb0478613b7027822502ef4f5aea671a8028d3aa5caf8bc4"},"schema_version":"1.0","source":{"id":"2607.03613","kind":"arxiv","version":1}},"canonical_sha256":"5aa3f5d3f80ca3b6b12002b350c16ae720f5d2c050dedcee49a49f04468252b4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5aa3f5d3f80ca3b6b12002b350c16ae720f5d2c050dedcee49a49f04468252b4","first_computed_at":"2026-07-07T02:17:57.470032Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-07T02:17:57.470032Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6ftq6wrNrPKrWxIQXMRT1GRplq/b3RhnyDDgm9Ne0ubLh5AmoLmvi92mh02d1jWnzAxy/X0TuLJETBnDFJDuCA==","signature_status":"signed_v1","signed_at":"2026-07-07T02:17:57.470798Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.03613","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4691b453af95b596bf6b1a0c1afc3196829286979a3654c2db6946d99ef33568","sha256:01b6b89f8a53e695c11a414d45b0736fa00d17f64c67982be09b823cc2f8a7e5"],"state_sha256":"8e1bfa0d797af12a1e038b12fc37c0ab3da90b7ca63e1ae68300ec3bcdcdc51e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DdMoq/s4S8W0V1udmCnX5u9H0ASVc/yY6qkOWWvi7yXIYaCCOItYzShAZUuj3ij/9vuSlmOwbd/ScXImMhrzCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T23:43:48.777958Z","bundle_sha256":"757240911108bbdacef8700b81bce0edc160ba961f8068240afda67cbf634943"}}