{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FVD2ZJGCRQXKFJXFVF3PISUK57","short_pith_number":"pith:FVD2ZJGC","schema_version":"1.0","canonical_sha256":"2d47aca4c28c2ea2a6e5a976f44a8aeff3a10183ead71267adfec006b627b69c","source":{"kind":"arxiv","id":"2507.05228","version":1},"attestation_state":"computed","paper":{"title":"Cascade: Token-Sharded Private LLM Inference","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Akilesh Potti, Arka Pal, Erica Choi, Louai Zahran, Micah Goldblum, Rahul Thomas","submitted_at":"2025-07-07T17:37:16Z","abstract_excerpt":"As LLMs continue to increase in parameter size, the computational resources required to run them are available to fewer parties. Therefore, third-party inference services -- where LLMs are hosted by third parties with significant computational resources -- are becoming increasingly popular. However, third party inference raises critical concerns about user data privacy. To mitigate these risks, privacy researchers have developed provably secure schemes for third-party inference, such as Secure Multi-Party Computation (SMPC). However, SMPC protocols have significant computational and communicat"},"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":"2507.05228","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-07T17:37:16Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"dc16598435f2eba50ccb44abd17e10aeeb5b34459f9a87029de6ec1622c17df2","abstract_canon_sha256":"557584fbe8e669749ddc1867c5d292e275c12d0e1080a3c11ddd17fe41503a86"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:33:10.871688Z","signature_b64":"jPke2pURRzZHMLDHQ7SJn7155p08fHW/x4WbqanXSQo2qB6QCxv4lWf2P8ouMr1mDj0+0YsJoNM144hCSJ3bBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2d47aca4c28c2ea2a6e5a976f44a8aeff3a10183ead71267adfec006b627b69c","last_reissued_at":"2026-07-05T11:33:10.871226Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:33:10.871226Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Cascade: Token-Sharded Private LLM Inference","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Akilesh Potti, Arka Pal, Erica Choi, Louai Zahran, Micah Goldblum, Rahul Thomas","submitted_at":"2025-07-07T17:37:16Z","abstract_excerpt":"As LLMs continue to increase in parameter size, the computational resources required to run them are available to fewer parties. Therefore, third-party inference services -- where LLMs are hosted by third parties with significant computational resources -- are becoming increasingly popular. However, third party inference raises critical concerns about user data privacy. To mitigate these risks, privacy researchers have developed provably secure schemes for third-party inference, such as Secure Multi-Party Computation (SMPC). However, SMPC protocols have significant computational and communicat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.05228","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/2507.05228/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":"2507.05228","created_at":"2026-07-05T11:33:10.871277+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.05228v1","created_at":"2026-07-05T11:33:10.871277+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.05228","created_at":"2026-07-05T11:33:10.871277+00:00"},{"alias_kind":"pith_short_12","alias_value":"FVD2ZJGCRQXK","created_at":"2026-07-05T11:33:10.871277+00:00"},{"alias_kind":"pith_short_16","alias_value":"FVD2ZJGCRQXKFJXF","created_at":"2026-07-05T11:33:10.871277+00:00"},{"alias_kind":"pith_short_8","alias_value":"FVD2ZJGC","created_at":"2026-07-05T11:33:10.871277+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2603.09002","citing_title":"Security Considerations for Multi-agent Systems","ref_index":253,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FVD2ZJGCRQXKFJXFVF3PISUK57","json":"https://pith.science/pith/FVD2ZJGCRQXKFJXFVF3PISUK57.json","graph_json":"https://pith.science/api/pith-number/FVD2ZJGCRQXKFJXFVF3PISUK57/graph.json","events_json":"https://pith.science/api/pith-number/FVD2ZJGCRQXKFJXFVF3PISUK57/events.json","paper":"https://pith.science/paper/FVD2ZJGC"},"agent_actions":{"view_html":"https://pith.science/pith/FVD2ZJGCRQXKFJXFVF3PISUK57","download_json":"https://pith.science/pith/FVD2ZJGCRQXKFJXFVF3PISUK57.json","view_paper":"https://pith.science/paper/FVD2ZJGC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.05228&json=true","fetch_graph":"https://pith.science/api/pith-number/FVD2ZJGCRQXKFJXFVF3PISUK57/graph.json","fetch_events":"https://pith.science/api/pith-number/FVD2ZJGCRQXKFJXFVF3PISUK57/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FVD2ZJGCRQXKFJXFVF3PISUK57/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FVD2ZJGCRQXKFJXFVF3PISUK57/action/storage_attestation","attest_author":"https://pith.science/pith/FVD2ZJGCRQXKFJXFVF3PISUK57/action/author_attestation","sign_citation":"https://pith.science/pith/FVD2ZJGCRQXKFJXFVF3PISUK57/action/citation_signature","submit_replication":"https://pith.science/pith/FVD2ZJGCRQXKFJXFVF3PISUK57/action/replication_record"}},"created_at":"2026-07-05T11:33:10.871277+00:00","updated_at":"2026-07-05T11:33:10.871277+00:00"}