{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:U2MKMRIROA4DXZ6JB3SIQWE3OC","short_pith_number":"pith:U2MKMRIR","schema_version":"1.0","canonical_sha256":"a698a6451170383be7c90ee488589b70a75dd6d435918e49b6f944f66639d777","source":{"kind":"arxiv","id":"2210.08674","version":1},"attestation_state":"computed","paper":{"title":"Scaling up Trustless DNN Inference with Zero-Knowledge Proofs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Daniel Kang, Ion Stoica, Tatsunori Hashimoto, Yi Sun","submitted_at":"2022-10-17T00:35:38Z","abstract_excerpt":"As ML models have increased in capabilities and accuracy, so has the complexity of their deployments. Increasingly, ML model consumers are turning to service providers to serve the ML models in the ML-as-a-service (MLaaS) paradigm. As MLaaS proliferates, a critical requirement emerges: how can model consumers verify that the correct predictions were served, in the face of malicious, lazy, or buggy service providers?\n  In this work, we present the first practical ImageNet-scale method to verify ML model inference non-interactively, i.e., after the inference has been done. To do so, we leverage "},"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":"2210.08674","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2022-10-17T00:35:38Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a82ebb6e1000e73fce2a707c50f1a8985c09808e832d080914712f6b6d7f7e42","abstract_canon_sha256":"e61c9ca8294cb672db52ba216f8c62a445334eb88954dac78344aa3e077ea205"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:07:13.519547Z","signature_b64":"3LbFHEWwJa5miiC20Oe7tNlWCvaYVABlQNVMrshqU8fk8mlZsnbMl9I20+zMnm5HO0hjBCU13HmSDctUFyCiBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a698a6451170383be7c90ee488589b70a75dd6d435918e49b6f944f66639d777","last_reissued_at":"2026-07-05T05:07:13.519044Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:07:13.519044Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scaling up Trustless DNN Inference with Zero-Knowledge Proofs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Daniel Kang, Ion Stoica, Tatsunori Hashimoto, Yi Sun","submitted_at":"2022-10-17T00:35:38Z","abstract_excerpt":"As ML models have increased in capabilities and accuracy, so has the complexity of their deployments. Increasingly, ML model consumers are turning to service providers to serve the ML models in the ML-as-a-service (MLaaS) paradigm. As MLaaS proliferates, a critical requirement emerges: how can model consumers verify that the correct predictions were served, in the face of malicious, lazy, or buggy service providers?\n  In this work, we present the first practical ImageNet-scale method to verify ML model inference non-interactively, i.e., after the inference has been done. To do so, we leverage "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.08674","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/2210.08674/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":"2210.08674","created_at":"2026-07-05T05:07:13.519100+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.08674v1","created_at":"2026-07-05T05:07:13.519100+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.08674","created_at":"2026-07-05T05:07:13.519100+00:00"},{"alias_kind":"pith_short_12","alias_value":"U2MKMRIROA4D","created_at":"2026-07-05T05:07:13.519100+00:00"},{"alias_kind":"pith_short_16","alias_value":"U2MKMRIROA4DXZ6J","created_at":"2026-07-05T05:07:13.519100+00:00"},{"alias_kind":"pith_short_8","alias_value":"U2MKMRIR","created_at":"2026-07-05T05:07:13.519100+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24942","citing_title":"Quantum-Resilient Decentralized AI Economies: Proof-of-Useful-Work and Post-Quantum Security","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04819","citing_title":"The Usefulness Gap in Proof-of-Useful-Work: An Empirical Study of Pearl's cuPOW Protocol","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05433","citing_title":"Zero knowledge verification for frontier AI training is possible","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21002","citing_title":"Verifiable Provenance and Watermarking for Generative AI: An Evidentiary Framework for International Operational Law and Domestic Courts","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2510.26576","citing_title":"\"Show Me You Comply... Without Showing Me Anything\": Zero-Knowledge Software Auditing for AI-Enabled Systems","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27666","citing_title":"VOW: Verifiable and Oblivious Watermark Detection for Large Language Models","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/U2MKMRIROA4DXZ6JB3SIQWE3OC","json":"https://pith.science/pith/U2MKMRIROA4DXZ6JB3SIQWE3OC.json","graph_json":"https://pith.science/api/pith-number/U2MKMRIROA4DXZ6JB3SIQWE3OC/graph.json","events_json":"https://pith.science/api/pith-number/U2MKMRIROA4DXZ6JB3SIQWE3OC/events.json","paper":"https://pith.science/paper/U2MKMRIR"},"agent_actions":{"view_html":"https://pith.science/pith/U2MKMRIROA4DXZ6JB3SIQWE3OC","download_json":"https://pith.science/pith/U2MKMRIROA4DXZ6JB3SIQWE3OC.json","view_paper":"https://pith.science/paper/U2MKMRIR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.08674&json=true","fetch_graph":"https://pith.science/api/pith-number/U2MKMRIROA4DXZ6JB3SIQWE3OC/graph.json","fetch_events":"https://pith.science/api/pith-number/U2MKMRIROA4DXZ6JB3SIQWE3OC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/U2MKMRIROA4DXZ6JB3SIQWE3OC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/U2MKMRIROA4DXZ6JB3SIQWE3OC/action/storage_attestation","attest_author":"https://pith.science/pith/U2MKMRIROA4DXZ6JB3SIQWE3OC/action/author_attestation","sign_citation":"https://pith.science/pith/U2MKMRIROA4DXZ6JB3SIQWE3OC/action/citation_signature","submit_replication":"https://pith.science/pith/U2MKMRIROA4DXZ6JB3SIQWE3OC/action/replication_record"}},"created_at":"2026-07-05T05:07:13.519100+00:00","updated_at":"2026-07-05T05:07:13.519100+00:00"}