{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:MT3GNTTENSY3VXWLXFT7BSF23P","short_pith_number":"pith:MT3GNTTE","schema_version":"1.0","canonical_sha256":"64f666ce646cb1badecbb967f0c8badbedf86b25e733cb3fafc16c9e88e3440f","source":{"kind":"arxiv","id":"2602.03812","version":2},"attestation_state":"computed","paper":{"title":"Antidistillation Fingerprinting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Alexander Robey, Asher Trockman, Fei Fang, John Kirchenbauer, J. Zico Kolter, Tom Goldstein, Yash Savani, Yixuan Even Xu","submitted_at":"2026-02-03T18:15:50Z","abstract_excerpt":"Model distillation enables efficient emulation of frontier large language models (LLMs), creating a need for robust mechanisms to detect when a third-party student model has trained on a teacher model's outputs. However, existing fingerprinting techniques that could be used to detect such distillation rely on heuristic perturbations that impose a steep trade-off between generation quality and fingerprinting strength, often requiring significant degradation of utility to ensure the fingerprint is effectively internalized by the student. We introduce antidistillation fingerprinting (ADFP), a pri"},"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":"2602.03812","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-02-03T18:15:50Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"451a01e0a0a01b3ebde368dbbc6e0b0ddec8e58d58751ee2b0864d90429d1466","abstract_canon_sha256":"1c3d2eeab3d98b524e0aa98eb39e01e6b3ee838aeb06133556391fcc21a3bc89"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:00:33.401878Z","signature_b64":"vIGwuSHNab8BnCW7fxHyCrUIPaWA21amMl3/fEU+FRnmNJ9/zMKkmMlQQQjslBVybENLM3UCo+NsB2N2RHHABw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"64f666ce646cb1badecbb967f0c8badbedf86b25e733cb3fafc16c9e88e3440f","last_reissued_at":"2026-05-20T00:00:33.401156Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:00:33.401156Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Antidistillation Fingerprinting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Alexander Robey, Asher Trockman, Fei Fang, John Kirchenbauer, J. Zico Kolter, Tom Goldstein, Yash Savani, Yixuan Even Xu","submitted_at":"2026-02-03T18:15:50Z","abstract_excerpt":"Model distillation enables efficient emulation of frontier large language models (LLMs), creating a need for robust mechanisms to detect when a third-party student model has trained on a teacher model's outputs. However, existing fingerprinting techniques that could be used to detect such distillation rely on heuristic perturbations that impose a steep trade-off between generation quality and fingerprinting strength, often requiring significant degradation of utility to ensure the fingerprint is effectively internalized by the student. We introduce antidistillation fingerprinting (ADFP), a pri"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.03812","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/2602.03812/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":"2602.03812","created_at":"2026-05-20T00:00:33.401278+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.03812v2","created_at":"2026-05-20T00:00:33.401278+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.03812","created_at":"2026-05-20T00:00:33.401278+00:00"},{"alias_kind":"pith_short_12","alias_value":"MT3GNTTENSY3","created_at":"2026-05-20T00:00:33.401278+00:00"},{"alias_kind":"pith_short_16","alias_value":"MT3GNTTENSY3VXWL","created_at":"2026-05-20T00:00:33.401278+00:00"},{"alias_kind":"pith_short_8","alias_value":"MT3GNTTE","created_at":"2026-05-20T00:00:33.401278+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2605.18829","citing_title":"Lossless Anti-Distillation Sampling","ref_index":131,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16462","citing_title":"Asking Back: Interaction-Layer Antidistillation Watermarks","ref_index":43,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MT3GNTTENSY3VXWLXFT7BSF23P","json":"https://pith.science/pith/MT3GNTTENSY3VXWLXFT7BSF23P.json","graph_json":"https://pith.science/api/pith-number/MT3GNTTENSY3VXWLXFT7BSF23P/graph.json","events_json":"https://pith.science/api/pith-number/MT3GNTTENSY3VXWLXFT7BSF23P/events.json","paper":"https://pith.science/paper/MT3GNTTE"},"agent_actions":{"view_html":"https://pith.science/pith/MT3GNTTENSY3VXWLXFT7BSF23P","download_json":"https://pith.science/pith/MT3GNTTENSY3VXWLXFT7BSF23P.json","view_paper":"https://pith.science/paper/MT3GNTTE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.03812&json=true","fetch_graph":"https://pith.science/api/pith-number/MT3GNTTENSY3VXWLXFT7BSF23P/graph.json","fetch_events":"https://pith.science/api/pith-number/MT3GNTTENSY3VXWLXFT7BSF23P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MT3GNTTENSY3VXWLXFT7BSF23P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MT3GNTTENSY3VXWLXFT7BSF23P/action/storage_attestation","attest_author":"https://pith.science/pith/MT3GNTTENSY3VXWLXFT7BSF23P/action/author_attestation","sign_citation":"https://pith.science/pith/MT3GNTTENSY3VXWLXFT7BSF23P/action/citation_signature","submit_replication":"https://pith.science/pith/MT3GNTTENSY3VXWLXFT7BSF23P/action/replication_record"}},"created_at":"2026-05-20T00:00:33.401278+00:00","updated_at":"2026-05-20T00:00:33.401278+00:00"}