{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KCMKHLUKL4KEVXYTVMPUACST34","short_pith_number":"pith:KCMKHLUK","schema_version":"1.0","canonical_sha256":"5098a3ae8a5f144adf13ab1f400a53df06d19e0a7e1d21930b87deab6c43f499","source":{"kind":"arxiv","id":"2509.24467","version":3},"attestation_state":"computed","paper":{"title":"Interpretable Self-Supervised Learning via Representer Landmarks and Nystr\\\"om Approximation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Debarghya Ghoshdastidar, Maedeh Zarvandi, Michael Timothy, Theresa Wasserer","submitted_at":"2025-09-29T08:45:40Z","abstract_excerpt":"Self-supervised learning (SSL) learns representations from massive unlabeled data, yet the resulting models typically operate as black boxes, necessitating domain-specific explanations. We introduce KREPES, a unified framework to analytically interpret the learned representations of SSL objectives, including SimCLR, BYOL, and VICReg. By bridging empirical neural tangent kernel approximations of neural networks with the Representer Theorem for kernels, we express the learned latent space directly via \"Representer Landmarks\", which are the representations of influential unlabeled training exampl"},"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":"2509.24467","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-09-29T08:45:40Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"2f7ae96c3030697b05e047996811c5942bf108b35f960337505e0de8accc3a01","abstract_canon_sha256":"e870ff06a5f9b913cc9dcf0321fa6b7ce9865e1d071b373d79ae45e99f6301bf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T02:04:47.949611Z","signature_b64":"LMU2HVDR4Tg5dCYgi68Ywlvvbsbv9prO1V8v2Q+jvyG4KN37xaHef6JWAim+Afi5d9gtdzRTXzK4s0jcw9oWBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5098a3ae8a5f144adf13ab1f400a53df06d19e0a7e1d21930b87deab6c43f499","last_reissued_at":"2026-06-02T02:04:47.949062Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T02:04:47.949062Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Interpretable Self-Supervised Learning via Representer Landmarks and Nystr\\\"om Approximation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Debarghya Ghoshdastidar, Maedeh Zarvandi, Michael Timothy, Theresa Wasserer","submitted_at":"2025-09-29T08:45:40Z","abstract_excerpt":"Self-supervised learning (SSL) learns representations from massive unlabeled data, yet the resulting models typically operate as black boxes, necessitating domain-specific explanations. We introduce KREPES, a unified framework to analytically interpret the learned representations of SSL objectives, including SimCLR, BYOL, and VICReg. By bridging empirical neural tangent kernel approximations of neural networks with the Representer Theorem for kernels, we express the learned latent space directly via \"Representer Landmarks\", which are the representations of influential unlabeled training exampl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.24467","kind":"arxiv","version":3},"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/2509.24467/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":"2509.24467","created_at":"2026-06-02T02:04:47.949126+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.24467v3","created_at":"2026-06-02T02:04:47.949126+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.24467","created_at":"2026-06-02T02:04:47.949126+00:00"},{"alias_kind":"pith_short_12","alias_value":"KCMKHLUKL4KE","created_at":"2026-06-02T02:04:47.949126+00:00"},{"alias_kind":"pith_short_16","alias_value":"KCMKHLUKL4KEVXYT","created_at":"2026-06-02T02:04:47.949126+00:00"},{"alias_kind":"pith_short_8","alias_value":"KCMKHLUK","created_at":"2026-06-02T02:04:47.949126+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KCMKHLUKL4KEVXYTVMPUACST34","json":"https://pith.science/pith/KCMKHLUKL4KEVXYTVMPUACST34.json","graph_json":"https://pith.science/api/pith-number/KCMKHLUKL4KEVXYTVMPUACST34/graph.json","events_json":"https://pith.science/api/pith-number/KCMKHLUKL4KEVXYTVMPUACST34/events.json","paper":"https://pith.science/paper/KCMKHLUK"},"agent_actions":{"view_html":"https://pith.science/pith/KCMKHLUKL4KEVXYTVMPUACST34","download_json":"https://pith.science/pith/KCMKHLUKL4KEVXYTVMPUACST34.json","view_paper":"https://pith.science/paper/KCMKHLUK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.24467&json=true","fetch_graph":"https://pith.science/api/pith-number/KCMKHLUKL4KEVXYTVMPUACST34/graph.json","fetch_events":"https://pith.science/api/pith-number/KCMKHLUKL4KEVXYTVMPUACST34/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KCMKHLUKL4KEVXYTVMPUACST34/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KCMKHLUKL4KEVXYTVMPUACST34/action/storage_attestation","attest_author":"https://pith.science/pith/KCMKHLUKL4KEVXYTVMPUACST34/action/author_attestation","sign_citation":"https://pith.science/pith/KCMKHLUKL4KEVXYTVMPUACST34/action/citation_signature","submit_replication":"https://pith.science/pith/KCMKHLUKL4KEVXYTVMPUACST34/action/replication_record"}},"created_at":"2026-06-02T02:04:47.949126+00:00","updated_at":"2026-06-02T02:04:47.949126+00:00"}