{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:2HQJIB4UM5UDL42SETHDFOL7KE","short_pith_number":"pith:2HQJIB4U","schema_version":"1.0","canonical_sha256":"d1e0940794676835f35224ce32b97f511c0aa35d4b279ce08b1572e2a2ab8276","source":{"kind":"arxiv","id":"2402.02675","version":2},"attestation_state":"computed","paper":{"title":"Verifiable evaluations of machine learning models using zkSNARKs","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.LG","authors_text":"Alexander Camuto, Alex 'Sandy' Pentland, Christian Paquin, Jason Morton, Robert Mahari, Shayla Nguyen, Shrey Jain, Tobin South","submitted_at":"2024-02-05T02:21:11Z","abstract_excerpt":"In a world of increasing closed-source commercial machine learning models, model evaluations from developers must be taken at face value. These benchmark results-whether over task accuracy, bias evaluations, or safety checks-are traditionally impossible to verify by a model end-user without the costly or impossible process of re-performing the benchmark on black-box model outputs. This work presents a method of verifiable model evaluation using model inference through zkSNARKs. The resulting zero-knowledge computational proofs of model outputs over datasets can be packaged into verifiable eval"},"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":"2402.02675","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-05T02:21:11Z","cross_cats_sorted":["cs.AI","cs.CR"],"title_canon_sha256":"da107074d63f3c299cdc74b17bf2d1d35f926ba37a627794ac0acc58660f6cbb","abstract_canon_sha256":"53ddf6eff15fd4e342edabcfd8bc36be008f5b5211ec62d7b76fd574387893cb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:21:49.409829Z","signature_b64":"8/QXU4dw/wzCMaLznBROwuSpCb8paChXeR5L/SXkKXeJdJl3M6k6ZSdxlXzbVULLPN00PQzy17jO64deS8KCCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d1e0940794676835f35224ce32b97f511c0aa35d4b279ce08b1572e2a2ab8276","last_reissued_at":"2026-07-05T08:21:49.409333Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:21:49.409333Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Verifiable evaluations of machine learning models using zkSNARKs","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.LG","authors_text":"Alexander Camuto, Alex 'Sandy' Pentland, Christian Paquin, Jason Morton, Robert Mahari, Shayla Nguyen, Shrey Jain, Tobin South","submitted_at":"2024-02-05T02:21:11Z","abstract_excerpt":"In a world of increasing closed-source commercial machine learning models, model evaluations from developers must be taken at face value. These benchmark results-whether over task accuracy, bias evaluations, or safety checks-are traditionally impossible to verify by a model end-user without the costly or impossible process of re-performing the benchmark on black-box model outputs. This work presents a method of verifiable model evaluation using model inference through zkSNARKs. The resulting zero-knowledge computational proofs of model outputs over datasets can be packaged into verifiable eval"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.02675","kind":"arxiv","version":2},"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/2402.02675/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":"2402.02675","created_at":"2026-07-05T08:21:49.409393+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.02675v2","created_at":"2026-07-05T08:21:49.409393+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.02675","created_at":"2026-07-05T08:21:49.409393+00:00"},{"alias_kind":"pith_short_12","alias_value":"2HQJIB4UM5UD","created_at":"2026-07-05T08:21:49.409393+00:00"},{"alias_kind":"pith_short_16","alias_value":"2HQJIB4UM5UDL42S","created_at":"2026-07-05T08:21:49.409393+00:00"},{"alias_kind":"pith_short_8","alias_value":"2HQJIB4U","created_at":"2026-07-05T08:21:49.409393+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09809","citing_title":"Evaluation Cards: An Interpretive Layer for AI Evaluation Reporting","ref_index":100,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05433","citing_title":"Zero knowledge verification for frontier AI training is possible","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04712","citing_title":"Hardware-Level Governance of AI Compute: A Feasibility Taxonomy for Regulatory Compliance and Treaty Verification","ref_index":48,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2HQJIB4UM5UDL42SETHDFOL7KE","json":"https://pith.science/pith/2HQJIB4UM5UDL42SETHDFOL7KE.json","graph_json":"https://pith.science/api/pith-number/2HQJIB4UM5UDL42SETHDFOL7KE/graph.json","events_json":"https://pith.science/api/pith-number/2HQJIB4UM5UDL42SETHDFOL7KE/events.json","paper":"https://pith.science/paper/2HQJIB4U"},"agent_actions":{"view_html":"https://pith.science/pith/2HQJIB4UM5UDL42SETHDFOL7KE","download_json":"https://pith.science/pith/2HQJIB4UM5UDL42SETHDFOL7KE.json","view_paper":"https://pith.science/paper/2HQJIB4U","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.02675&json=true","fetch_graph":"https://pith.science/api/pith-number/2HQJIB4UM5UDL42SETHDFOL7KE/graph.json","fetch_events":"https://pith.science/api/pith-number/2HQJIB4UM5UDL42SETHDFOL7KE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2HQJIB4UM5UDL42SETHDFOL7KE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2HQJIB4UM5UDL42SETHDFOL7KE/action/storage_attestation","attest_author":"https://pith.science/pith/2HQJIB4UM5UDL42SETHDFOL7KE/action/author_attestation","sign_citation":"https://pith.science/pith/2HQJIB4UM5UDL42SETHDFOL7KE/action/citation_signature","submit_replication":"https://pith.science/pith/2HQJIB4UM5UDL42SETHDFOL7KE/action/replication_record"}},"created_at":"2026-07-05T08:21:49.409393+00:00","updated_at":"2026-07-05T08:21:49.409393+00:00"}