{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:YFMWH5XGSQWWVVALJFL2U2G2FX","short_pith_number":"pith:YFMWH5XG","canonical_record":{"source":{"id":"2602.10217","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-10T19:08:40Z","cross_cats_sorted":[],"title_canon_sha256":"12dd06c0100b3d201edd00d0bf168a0c7de9ae166ed243aa99bae3682d1c1a89","abstract_canon_sha256":"a46d3d40c818fd3e991b329e36383d97c2bd0ac1704cf3a00b9818d5e1d86eed"},"schema_version":"1.0"},"canonical_sha256":"c15963f6e6942d6ad40b4957aa68da2dca98847284d2b94e9be3263315671e15","source":{"kind":"arxiv","id":"2602.10217","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.10217","created_at":"2026-06-23T01:13:03Z"},{"alias_kind":"arxiv_version","alias_value":"2602.10217v2","created_at":"2026-06-23T01:13:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.10217","created_at":"2026-06-23T01:13:03Z"},{"alias_kind":"pith_short_12","alias_value":"YFMWH5XGSQWW","created_at":"2026-06-23T01:13:03Z"},{"alias_kind":"pith_short_16","alias_value":"YFMWH5XGSQWWVVAL","created_at":"2026-06-23T01:13:03Z"},{"alias_kind":"pith_short_8","alias_value":"YFMWH5XG","created_at":"2026-06-23T01:13:03Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:YFMWH5XGSQWWVVALJFL2U2G2FX","target":"record","payload":{"canonical_record":{"source":{"id":"2602.10217","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-10T19:08:40Z","cross_cats_sorted":[],"title_canon_sha256":"12dd06c0100b3d201edd00d0bf168a0c7de9ae166ed243aa99bae3682d1c1a89","abstract_canon_sha256":"a46d3d40c818fd3e991b329e36383d97c2bd0ac1704cf3a00b9818d5e1d86eed"},"schema_version":"1.0"},"canonical_sha256":"c15963f6e6942d6ad40b4957aa68da2dca98847284d2b94e9be3263315671e15","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:13:03.527226Z","signature_b64":"2j+qTGr4tgYFmeiIWIawetB6xu+b+5XhYKLtRfLpDQ45Fpkab+3foKQntd+Q1etLaMeU4uNdIltwMLfkjLJjCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c15963f6e6942d6ad40b4957aa68da2dca98847284d2b94e9be3263315671e15","last_reissued_at":"2026-06-23T01:13:03.526660Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:13:03.526660Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2602.10217","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-23T01:13:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QvIYZnOxHsOF+nxaZmHwh8yfKvVrP1K6cZaQRXqDPS8jGabsb/j6ZhomVFcgR0zxy/uGVV8sVZalDZn8fp2rCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T18:02:51.040629Z"},"content_sha256":"53d5360973b680af91fb64b84d8d876b1089eec3865f53aabd66d29658829234","schema_version":"1.0","event_id":"sha256:53d5360973b680af91fb64b84d8d876b1089eec3865f53aabd66d29658829234"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:YFMWH5XGSQWWVVALJFL2U2G2FX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Temper-Then-Tilt: Principled Unlearning for Generative Models through Tempering and Classifier Guidance","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aryan Mokhtari, Jacob L. Block, Mehryar Mohri, Sanjay Shakkottai","submitted_at":"2026-02-10T19:08:40Z","abstract_excerpt":"We study machine unlearning in large generative models by framing the task as density ratio estimation to a target distribution rather than supervised fine-tuning. While classifier guidance is a standard approach for approximating this ratio and can succeed in general, we show it can fail to faithfully unlearn with finite samples when the forget set represents a sharp, concentrated data distribution. To address this, we introduce Temper-Then-Tilt Unlearning (T3-Unlearning), which freezes the base model and applies a two-step inference procedure: (i) tempering the base distribution to flatten h"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.10217","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/2602.10217/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-23T01:13:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xjyuzV0eB+9wftA3XiYdxZNV+VCzj9q0S/TuUC0mCLPfB/noMiZMWq23JW003XM2/xhWe7lVtnXMLR9m6AWhDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T18:02:51.041213Z"},"content_sha256":"95f7970a125b8a8441625b56c00aaca6efac2f4ddfe51a781e219d9b74f7bb9c","schema_version":"1.0","event_id":"sha256:95f7970a125b8a8441625b56c00aaca6efac2f4ddfe51a781e219d9b74f7bb9c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YFMWH5XGSQWWVVALJFL2U2G2FX/bundle.json","state_url":"https://pith.science/pith/YFMWH5XGSQWWVVALJFL2U2G2FX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YFMWH5XGSQWWVVALJFL2U2G2FX/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-06T18:02:51Z","links":{"resolver":"https://pith.science/pith/YFMWH5XGSQWWVVALJFL2U2G2FX","bundle":"https://pith.science/pith/YFMWH5XGSQWWVVALJFL2U2G2FX/bundle.json","state":"https://pith.science/pith/YFMWH5XGSQWWVVALJFL2U2G2FX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YFMWH5XGSQWWVVALJFL2U2G2FX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:YFMWH5XGSQWWVVALJFL2U2G2FX","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"a46d3d40c818fd3e991b329e36383d97c2bd0ac1704cf3a00b9818d5e1d86eed","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-10T19:08:40Z","title_canon_sha256":"12dd06c0100b3d201edd00d0bf168a0c7de9ae166ed243aa99bae3682d1c1a89"},"schema_version":"1.0","source":{"id":"2602.10217","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.10217","created_at":"2026-06-23T01:13:03Z"},{"alias_kind":"arxiv_version","alias_value":"2602.10217v2","created_at":"2026-06-23T01:13:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.10217","created_at":"2026-06-23T01:13:03Z"},{"alias_kind":"pith_short_12","alias_value":"YFMWH5XGSQWW","created_at":"2026-06-23T01:13:03Z"},{"alias_kind":"pith_short_16","alias_value":"YFMWH5XGSQWWVVAL","created_at":"2026-06-23T01:13:03Z"},{"alias_kind":"pith_short_8","alias_value":"YFMWH5XG","created_at":"2026-06-23T01:13:03Z"}],"graph_snapshots":[{"event_id":"sha256:95f7970a125b8a8441625b56c00aaca6efac2f4ddfe51a781e219d9b74f7bb9c","target":"graph","created_at":"2026-06-23T01:13:03Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2602.10217/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We study machine unlearning in large generative models by framing the task as density ratio estimation to a target distribution rather than supervised fine-tuning. While classifier guidance is a standard approach for approximating this ratio and can succeed in general, we show it can fail to faithfully unlearn with finite samples when the forget set represents a sharp, concentrated data distribution. To address this, we introduce Temper-Then-Tilt Unlearning (T3-Unlearning), which freezes the base model and applies a two-step inference procedure: (i) tempering the base distribution to flatten h","authors_text":"Aryan Mokhtari, Jacob L. Block, Mehryar Mohri, Sanjay Shakkottai","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-10T19:08:40Z","title":"Temper-Then-Tilt: Principled Unlearning for Generative Models through Tempering and Classifier Guidance"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.10217","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:53d5360973b680af91fb64b84d8d876b1089eec3865f53aabd66d29658829234","target":"record","created_at":"2026-06-23T01:13:03Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"a46d3d40c818fd3e991b329e36383d97c2bd0ac1704cf3a00b9818d5e1d86eed","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-10T19:08:40Z","title_canon_sha256":"12dd06c0100b3d201edd00d0bf168a0c7de9ae166ed243aa99bae3682d1c1a89"},"schema_version":"1.0","source":{"id":"2602.10217","kind":"arxiv","version":2}},"canonical_sha256":"c15963f6e6942d6ad40b4957aa68da2dca98847284d2b94e9be3263315671e15","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c15963f6e6942d6ad40b4957aa68da2dca98847284d2b94e9be3263315671e15","first_computed_at":"2026-06-23T01:13:03.526660Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-23T01:13:03.526660Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2j+qTGr4tgYFmeiIWIawetB6xu+b+5XhYKLtRfLpDQ45Fpkab+3foKQntd+Q1etLaMeU4uNdIltwMLfkjLJjCA==","signature_status":"signed_v1","signed_at":"2026-06-23T01:13:03.527226Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.10217","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:53d5360973b680af91fb64b84d8d876b1089eec3865f53aabd66d29658829234","sha256:95f7970a125b8a8441625b56c00aaca6efac2f4ddfe51a781e219d9b74f7bb9c"],"state_sha256":"23b4843f5d1ba6cd87a5855e5750cfa5b25878990d0035db53e13bf1522ce361"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UZ71b3m1T3IPoSrX6MmdS5VGLhjAktAvRR4FIIFx1qEeU5c/u/w9Mp9EX49Xm/Hp0bD1RBtJFHnng7f5Ag29Bg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T18:02:51.045709Z","bundle_sha256":"25427735ed8b53a4295b9c35f30061da03fdf5f7319a7da2282280dff2adda4d"}}