{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:QHCRETQOTIDITLRT4CS3H2MAWL","short_pith_number":"pith:QHCRETQO","schema_version":"1.0","canonical_sha256":"81c5124e0e9a0689ae33e0a5b3e980b2c2994873869bbe998e071532229951a9","source":{"kind":"arxiv","id":"2308.02760","version":1},"attestation_state":"computed","paper":{"title":"Neural Collapse in the Intermediate Hidden Layers of Classification Neural Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Anton Stengel, Emre Onal, Jake Intrater, Liam Parker","submitted_at":"2023-08-05T01:19:38Z","abstract_excerpt":"Neural Collapse (NC) gives a precise description of the representations of classes in the final hidden layer of classification neural networks. This description provides insights into how these networks learn features and generalize well when trained past zero training error. However, to date, (NC) has only been studied in the final layer of these networks. In the present paper, we provide the first comprehensive empirical analysis of the emergence of (NC) in the intermediate hidden layers of these classifiers. We examine a variety of network architectures, activations, and datasets, and demon"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2308.02760","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-08-05T01:19:38Z","cross_cats_sorted":[],"title_canon_sha256":"c22f17e54b1582d386bd7e211c664a31754480e79e732f509917f7ac4fd95669","abstract_canon_sha256":"f2a49e0685adc64528fc39cc2be903786b667b1fc89ccb0735bb679cffb6dae8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:38:06.971863Z","signature_b64":"Eo/bTfoXwKHrKr2z+0HzeF6RDo6INOMs3ovei3hHeaG232K4OwdP3WUOyGg+2RnFDT4//XLmsbh3wdOUzOoVDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"81c5124e0e9a0689ae33e0a5b3e980b2c2994873869bbe998e071532229951a9","last_reissued_at":"2026-07-05T06:38:06.971403Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:38:06.971403Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Neural Collapse in the Intermediate Hidden Layers of Classification Neural Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Anton Stengel, Emre Onal, Jake Intrater, Liam Parker","submitted_at":"2023-08-05T01:19:38Z","abstract_excerpt":"Neural Collapse (NC) gives a precise description of the representations of classes in the final hidden layer of classification neural networks. This description provides insights into how these networks learn features and generalize well when trained past zero training error. However, to date, (NC) has only been studied in the final layer of these networks. In the present paper, we provide the first comprehensive empirical analysis of the emergence of (NC) in the intermediate hidden layers of these classifiers. We examine a variety of network architectures, activations, and datasets, and demon"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.02760","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/2308.02760/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2308.02760","created_at":"2026-07-05T06:38:06.971462+00:00"},{"alias_kind":"arxiv_version","alias_value":"2308.02760v1","created_at":"2026-07-05T06:38:06.971462+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.02760","created_at":"2026-07-05T06:38:06.971462+00:00"},{"alias_kind":"pith_short_12","alias_value":"QHCRETQOTIDI","created_at":"2026-07-05T06:38:06.971462+00:00"},{"alias_kind":"pith_short_16","alias_value":"QHCRETQOTIDITLRT","created_at":"2026-07-05T06:38:06.971462+00:00"},{"alias_kind":"pith_short_8","alias_value":"QHCRETQO","created_at":"2026-07-05T06:38:06.971462+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.18598","citing_title":"Pointwise Generalization in Deep Neural Networks","ref_index":94,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QHCRETQOTIDITLRT4CS3H2MAWL","json":"https://pith.science/pith/QHCRETQOTIDITLRT4CS3H2MAWL.json","graph_json":"https://pith.science/api/pith-number/QHCRETQOTIDITLRT4CS3H2MAWL/graph.json","events_json":"https://pith.science/api/pith-number/QHCRETQOTIDITLRT4CS3H2MAWL/events.json","paper":"https://pith.science/paper/QHCRETQO"},"agent_actions":{"view_html":"https://pith.science/pith/QHCRETQOTIDITLRT4CS3H2MAWL","download_json":"https://pith.science/pith/QHCRETQOTIDITLRT4CS3H2MAWL.json","view_paper":"https://pith.science/paper/QHCRETQO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2308.02760&json=true","fetch_graph":"https://pith.science/api/pith-number/QHCRETQOTIDITLRT4CS3H2MAWL/graph.json","fetch_events":"https://pith.science/api/pith-number/QHCRETQOTIDITLRT4CS3H2MAWL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QHCRETQOTIDITLRT4CS3H2MAWL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QHCRETQOTIDITLRT4CS3H2MAWL/action/storage_attestation","attest_author":"https://pith.science/pith/QHCRETQOTIDITLRT4CS3H2MAWL/action/author_attestation","sign_citation":"https://pith.science/pith/QHCRETQOTIDITLRT4CS3H2MAWL/action/citation_signature","submit_replication":"https://pith.science/pith/QHCRETQOTIDITLRT4CS3H2MAWL/action/replication_record"}},"created_at":"2026-07-05T06:38:06.971462+00:00","updated_at":"2026-07-05T06:38:06.971462+00:00"}