{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:UKJ444OWOR2LQMZ6MATEYVMBN4","short_pith_number":"pith:UKJ444OW","canonical_record":{"source":{"id":"2503.12347","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-16T04:00:32Z","cross_cats_sorted":[],"title_canon_sha256":"164ed02d10456f00035bbaf94e01877af20a1b6811a8cc8eb96a3aab35d65155","abstract_canon_sha256":"4f18f0958de090dd3b146fed84544dfdf6ea040dd9a27fbb47d46beccf351192"},"schema_version":"1.0"},"canonical_sha256":"a293ce71d67474b8333e60264c55816f296be3a1275d354e7a874353b9fe025a","source":{"kind":"arxiv","id":"2503.12347","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.12347","created_at":"2026-07-05T11:38:28Z"},{"alias_kind":"arxiv_version","alias_value":"2503.12347v2","created_at":"2026-07-05T11:38:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.12347","created_at":"2026-07-05T11:38:28Z"},{"alias_kind":"pith_short_12","alias_value":"UKJ444OWOR2L","created_at":"2026-07-05T11:38:28Z"},{"alias_kind":"pith_short_16","alias_value":"UKJ444OWOR2LQMZ6","created_at":"2026-07-05T11:38:28Z"},{"alias_kind":"pith_short_8","alias_value":"UKJ444OW","created_at":"2026-07-05T11:38:28Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:UKJ444OWOR2LQMZ6MATEYVMBN4","target":"record","payload":{"canonical_record":{"source":{"id":"2503.12347","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-16T04:00:32Z","cross_cats_sorted":[],"title_canon_sha256":"164ed02d10456f00035bbaf94e01877af20a1b6811a8cc8eb96a3aab35d65155","abstract_canon_sha256":"4f18f0958de090dd3b146fed84544dfdf6ea040dd9a27fbb47d46beccf351192"},"schema_version":"1.0"},"canonical_sha256":"a293ce71d67474b8333e60264c55816f296be3a1275d354e7a874353b9fe025a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:38:28.008057Z","signature_b64":"j8s5pyZy4bzt2QuUbJXzwLZPNT4XxE4qgA98zvGiT5c2Aejt4EYkn4evG3/pQRrSVaUKp4PZCuwXf0WEZkvlBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a293ce71d67474b8333e60264c55816f296be3a1275d354e7a874353b9fe025a","last_reissued_at":"2026-07-05T11:38:28.007527Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:38:28.007527Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.12347","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":"2026-07-05T11:38:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TIVdBreEsKQVVMV2epeX7Wl0OQLZ8s3ESa/d7FxBIqEN2JJKUNqdav+gE3hFXBZjmyZt9c6lXpdDEhRTgibnBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T07:18:54.335136Z"},"content_sha256":"5440230a3922d9314f0a79a8a14d78916730b91872c1fb67954a407402a1c9bd","schema_version":"1.0","event_id":"sha256:5440230a3922d9314f0a79a8a14d78916730b91872c1fb67954a407402a1c9bd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:UKJ444OWOR2LQMZ6MATEYVMBN4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Synthesizing Privacy-Preserving Text Data via Finetuning without Finetuning Billion-Scale LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bowen Tan, Eric Xing, Shanshan Wu, Zheng Xu, Zhiting Hu","submitted_at":"2025-03-16T04:00:32Z","abstract_excerpt":"Synthetic data offers a promising path to train models while preserving data privacy. Differentially private (DP) finetuning of large language models (LLMs) as data generator is effective, but is impractical when computation resources are limited. Meanwhile, prompt-based methods such as private evolution depend heavily on the manual prompts, and ineffectively use private information in their iterative data selection process. To overcome these limitations, we propose CTCL (Data Synthesis with ConTrollability and CLustering), a novel framework for generating privacy-preserving synthetic data wit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.12347","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/2503.12347/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":"2026-07-05T11:38:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fmNaJ/SKFm3pDXMMt9cNX0yT0dDJmpJ8/M3Q4rp3RxEv5QEsFvYyaKQumozKflQcZWJg1Kmtr+hdAyrxLqcYAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T07:18:54.335658Z"},"content_sha256":"58836fb26c82539f5e2625b3aea85f5c37fa2dfd36b69b5c81bb4b589b388ebc","schema_version":"1.0","event_id":"sha256:58836fb26c82539f5e2625b3aea85f5c37fa2dfd36b69b5c81bb4b589b388ebc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UKJ444OWOR2LQMZ6MATEYVMBN4/bundle.json","state_url":"https://pith.science/pith/UKJ444OWOR2LQMZ6MATEYVMBN4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UKJ444OWOR2LQMZ6MATEYVMBN4/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":"2026-08-08T07:18:54Z","links":{"resolver":"https://pith.science/pith/UKJ444OWOR2LQMZ6MATEYVMBN4","bundle":"https://pith.science/pith/UKJ444OWOR2LQMZ6MATEYVMBN4/bundle.json","state":"https://pith.science/pith/UKJ444OWOR2LQMZ6MATEYVMBN4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UKJ444OWOR2LQMZ6MATEYVMBN4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:UKJ444OWOR2LQMZ6MATEYVMBN4","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":"4f18f0958de090dd3b146fed84544dfdf6ea040dd9a27fbb47d46beccf351192","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-16T04:00:32Z","title_canon_sha256":"164ed02d10456f00035bbaf94e01877af20a1b6811a8cc8eb96a3aab35d65155"},"schema_version":"1.0","source":{"id":"2503.12347","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.12347","created_at":"2026-07-05T11:38:28Z"},{"alias_kind":"arxiv_version","alias_value":"2503.12347v2","created_at":"2026-07-05T11:38:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.12347","created_at":"2026-07-05T11:38:28Z"},{"alias_kind":"pith_short_12","alias_value":"UKJ444OWOR2L","created_at":"2026-07-05T11:38:28Z"},{"alias_kind":"pith_short_16","alias_value":"UKJ444OWOR2LQMZ6","created_at":"2026-07-05T11:38:28Z"},{"alias_kind":"pith_short_8","alias_value":"UKJ444OW","created_at":"2026-07-05T11:38:28Z"}],"graph_snapshots":[{"event_id":"sha256:58836fb26c82539f5e2625b3aea85f5c37fa2dfd36b69b5c81bb4b589b388ebc","target":"graph","created_at":"2026-07-05T11:38:28Z","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/2503.12347/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Synthetic data offers a promising path to train models while preserving data privacy. Differentially private (DP) finetuning of large language models (LLMs) as data generator is effective, but is impractical when computation resources are limited. Meanwhile, prompt-based methods such as private evolution depend heavily on the manual prompts, and ineffectively use private information in their iterative data selection process. To overcome these limitations, we propose CTCL (Data Synthesis with ConTrollability and CLustering), a novel framework for generating privacy-preserving synthetic data wit","authors_text":"Bowen Tan, Eric Xing, Shanshan Wu, Zheng Xu, Zhiting Hu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-16T04:00:32Z","title":"Synthesizing Privacy-Preserving Text Data via Finetuning without Finetuning Billion-Scale LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.12347","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":"sha256:5440230a3922d9314f0a79a8a14d78916730b91872c1fb67954a407402a1c9bd","target":"record","created_at":"2026-07-05T11:38:28Z","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":"4f18f0958de090dd3b146fed84544dfdf6ea040dd9a27fbb47d46beccf351192","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-16T04:00:32Z","title_canon_sha256":"164ed02d10456f00035bbaf94e01877af20a1b6811a8cc8eb96a3aab35d65155"},"schema_version":"1.0","source":{"id":"2503.12347","kind":"arxiv","version":2}},"canonical_sha256":"a293ce71d67474b8333e60264c55816f296be3a1275d354e7a874353b9fe025a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a293ce71d67474b8333e60264c55816f296be3a1275d354e7a874353b9fe025a","first_computed_at":"2026-07-05T11:38:28.007527Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:38:28.007527Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"j8s5pyZy4bzt2QuUbJXzwLZPNT4XxE4qgA98zvGiT5c2Aejt4EYkn4evG3/pQRrSVaUKp4PZCuwXf0WEZkvlBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:38:28.008057Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.12347","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5440230a3922d9314f0a79a8a14d78916730b91872c1fb67954a407402a1c9bd","sha256:58836fb26c82539f5e2625b3aea85f5c37fa2dfd36b69b5c81bb4b589b388ebc"],"state_sha256":"29c04290609606f790bf0a760b09b204cfab158f83dfe82daf36adaf05df5036"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1ZRttAalDXgkCXGNQRcqd6e89m+I5Gc03TibMmWz1+JUoFMhpdee0OWMNTaysxQg5MpbAirxPh7W2q3m+pmDAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T07:18:54.339517Z","bundle_sha256":"416f452488eab0395d8657ea24faa2be3a9ac9aa51fa01340048c82c3d4baae9"}}