{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:R6X2YCE4THZDXIGY7KRIVD6T4K","short_pith_number":"pith:R6X2YCE4","schema_version":"1.0","canonical_sha256":"8fafac089c99f23ba0d8faa28a8fd3e2935129ee39ef20962b9a4066993937cd","source":{"kind":"arxiv","id":"2606.20347","version":1},"attestation_state":"computed","paper":{"title":"Critical Percolation as a Synthetic Data Model for Interpretability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn"],"primary_cat":"cs.LG","authors_text":"Aryeh Brill, Tom Ingebretsen Carlson","submitted_at":"2026-06-18T15:15:57Z","abstract_excerpt":"Neural networks learn features that reflect the hierarchical, multi-scale structure of natural data. Synthetic datasets used to evaluate interpretability methods typically lack this structure, limiting their value as realistic toy models. To close this gap, we introduce a family of synthetic datasets consisting of hierarchical functions defined on critical mean-field percolation clusters embedded in a high-dimensional data space. The percolation data consists of sparse, low-dimensional fractal clusters with a power-law size distribution. Latent variables modeling a taxonomic hierarchy generate"},"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":"2606.20347","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-18T15:15:57Z","cross_cats_sorted":["cond-mat.dis-nn"],"title_canon_sha256":"2934368f1ac67b3e49e98e3c80dba5a25f3e2d41d50e6fb5afb1853f37f62378","abstract_canon_sha256":"3af7d9d032aa6c998fff3822496412fa852ee776fdb99c4605275b0bdc17ed86"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:13:09.520106Z","signature_b64":"SptaJwyDDORygTfkIigsfNnh7wRjFalgpxkI6LyVsTIG7FfG/UGtULPq5ZH+QppHylD0Ce91wU9Qldlf0UcjCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8fafac089c99f23ba0d8faa28a8fd3e2935129ee39ef20962b9a4066993937cd","last_reissued_at":"2026-06-19T16:13:09.519757Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:13:09.519757Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Critical Percolation as a Synthetic Data Model for Interpretability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn"],"primary_cat":"cs.LG","authors_text":"Aryeh Brill, Tom Ingebretsen Carlson","submitted_at":"2026-06-18T15:15:57Z","abstract_excerpt":"Neural networks learn features that reflect the hierarchical, multi-scale structure of natural data. Synthetic datasets used to evaluate interpretability methods typically lack this structure, limiting their value as realistic toy models. To close this gap, we introduce a family of synthetic datasets consisting of hierarchical functions defined on critical mean-field percolation clusters embedded in a high-dimensional data space. The percolation data consists of sparse, low-dimensional fractal clusters with a power-law size distribution. Latent variables modeling a taxonomic hierarchy generate"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20347","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/2606.20347/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":"2606.20347","created_at":"2026-06-19T16:13:09.519823+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.20347v1","created_at":"2026-06-19T16:13:09.519823+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20347","created_at":"2026-06-19T16:13:09.519823+00:00"},{"alias_kind":"pith_short_12","alias_value":"R6X2YCE4THZD","created_at":"2026-06-19T16:13:09.519823+00:00"},{"alias_kind":"pith_short_16","alias_value":"R6X2YCE4THZDXIGY","created_at":"2026-06-19T16:13:09.519823+00:00"},{"alias_kind":"pith_short_8","alias_value":"R6X2YCE4","created_at":"2026-06-19T16:13:09.519823+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/R6X2YCE4THZDXIGY7KRIVD6T4K","json":"https://pith.science/pith/R6X2YCE4THZDXIGY7KRIVD6T4K.json","graph_json":"https://pith.science/api/pith-number/R6X2YCE4THZDXIGY7KRIVD6T4K/graph.json","events_json":"https://pith.science/api/pith-number/R6X2YCE4THZDXIGY7KRIVD6T4K/events.json","paper":"https://pith.science/paper/R6X2YCE4"},"agent_actions":{"view_html":"https://pith.science/pith/R6X2YCE4THZDXIGY7KRIVD6T4K","download_json":"https://pith.science/pith/R6X2YCE4THZDXIGY7KRIVD6T4K.json","view_paper":"https://pith.science/paper/R6X2YCE4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.20347&json=true","fetch_graph":"https://pith.science/api/pith-number/R6X2YCE4THZDXIGY7KRIVD6T4K/graph.json","fetch_events":"https://pith.science/api/pith-number/R6X2YCE4THZDXIGY7KRIVD6T4K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R6X2YCE4THZDXIGY7KRIVD6T4K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R6X2YCE4THZDXIGY7KRIVD6T4K/action/storage_attestation","attest_author":"https://pith.science/pith/R6X2YCE4THZDXIGY7KRIVD6T4K/action/author_attestation","sign_citation":"https://pith.science/pith/R6X2YCE4THZDXIGY7KRIVD6T4K/action/citation_signature","submit_replication":"https://pith.science/pith/R6X2YCE4THZDXIGY7KRIVD6T4K/action/replication_record"}},"created_at":"2026-06-19T16:13:09.519823+00:00","updated_at":"2026-06-19T16:13:09.519823+00:00"}