{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LPH6T5NM6EZ3BVPTSI7NPVSVZ6","short_pith_number":"pith:LPH6T5NM","schema_version":"1.0","canonical_sha256":"5bcfe9f5acf133b0d5f3923ed7d655cfbddc63545e95bdba97fcc61944227ad5","source":{"kind":"arxiv","id":"2605.01735","version":2},"attestation_state":"computed","paper":{"title":"Less is More: Geometric Unlearning for LLMs with Minimal Data Disclosure","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Geometric Unlearning lets LLMs forget specific private facts using only a handful of synthetic prompts while retaining general performance.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chenchen Tan, Cunjian Chen, Longxiang Gao, Shujie Cui, Xinghao Li, Youyang Qu","submitted_at":"2026-05-03T06:20:03Z","abstract_excerpt":"As large language models (LLMs) are increasingly deployed in real-world systems, they must support post-hoc removal of specific content to meet privacy and governance requirements. This motivates selective unlearning, which suppresses information about a particular entity or topic while preserving the LLM's general utility. However, most existing LLM unlearning methods require access to the original training corpus and rely on output-level refusal tuning or broad gradient updates, creating a tension among unlearning strength, non-target preservation, and data availability. We propose Geometric"},"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":"2605.01735","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-03T06:20:03Z","cross_cats_sorted":[],"title_canon_sha256":"d0dca146ca6ae22b09ad7a3a1b68227629cf8b01ce07edf5f79e8523fbec5cb9","abstract_canon_sha256":"cd49f95c3619721b41c2f54a7b45db22a147bc8f16cb6f123b2c130f4fe60b06"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:05:12.458057Z","signature_b64":"u7JE9+Hean6Kt2shkiRLsJCrvbK9Cm3kkc4sYrt1eE1Cyq3YB11oqUXjvgLHHxPtKswN5qmHAAZX5Xxmv0iSAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5bcfe9f5acf133b0d5f3923ed7d655cfbddc63545e95bdba97fcc61944227ad5","last_reissued_at":"2026-05-28T01:05:12.457391Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:05:12.457391Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Less is More: Geometric Unlearning for LLMs with Minimal Data Disclosure","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Geometric Unlearning lets LLMs forget specific private facts using only a handful of synthetic prompts while retaining general performance.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chenchen Tan, Cunjian Chen, Longxiang Gao, Shujie Cui, Xinghao Li, Youyang Qu","submitted_at":"2026-05-03T06:20:03Z","abstract_excerpt":"As large language models (LLMs) are increasingly deployed in real-world systems, they must support post-hoc removal of specific content to meet privacy and governance requirements. This motivates selective unlearning, which suppresses information about a particular entity or topic while preserving the LLM's general utility. However, most existing LLM unlearning methods require access to the original training corpus and rely on output-level refusal tuning or broad gradient updates, creating a tension among unlearning strength, non-target preservation, and data availability. We propose Geometric"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Across privacy-oriented unlearning benchmarks (ToFU and UnlearnPII), GU achieves strong target suppression with minimal impact on non-target performance, demonstrating that effective unlearning can be achieved with minimal synthetic data.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The assumption that a compact low-rank geometry distilled from safe reference prompts, combined with projection-based alignment using synthetic anchors, can effectively suppress target information in the model's hidden states without access to the original training corpus or significant utility loss.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Geometric Unlearning suppresses specific knowledge in LLMs by projecting hidden planning states onto a low-rank safe geometry derived from minimal reference prompts.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Geometric Unlearning lets LLMs forget specific private facts using only a handful of synthetic prompts while retaining general performance.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"60b0f8712d9d23ca96357302aab8f3362d99a638b9a2f4d51e0250f8b44530c0"},"source":{"id":"2605.01735","kind":"arxiv","version":2},"verdict":{"id":"504d7cdf-3774-41e3-b5cf-790afc4c786a","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T15:58:11.548449Z","strongest_claim":"Across privacy-oriented unlearning benchmarks (ToFU and UnlearnPII), GU achieves strong target suppression with minimal impact on non-target performance, demonstrating that effective unlearning can be achieved with minimal synthetic data.","one_line_summary":"Geometric Unlearning suppresses specific knowledge in LLMs by projecting hidden planning states onto a low-rank safe geometry derived from minimal reference prompts.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The assumption that a compact low-rank geometry distilled from safe reference prompts, combined with projection-based alignment using synthetic anchors, can effectively suppress target information in the model's hidden states without access to the original training corpus or significant utility loss.","pith_extraction_headline":"Geometric Unlearning lets LLMs forget specific private facts using only a handful of synthetic prompts while retaining general performance."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.01735/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T17:37:19.650148Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-20T05:01:23.091965Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T17:00:50.088978Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"2792726b88a81447be03310858e3b22ff01e6d40439872c768522f65e9d3fe66"},"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":"2605.01735","created_at":"2026-05-28T01:05:12.457524+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.01735v2","created_at":"2026-05-28T01:05:12.457524+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.01735","created_at":"2026-05-28T01:05:12.457524+00:00"},{"alias_kind":"pith_short_12","alias_value":"LPH6T5NM6EZ3","created_at":"2026-05-28T01:05:12.457524+00:00"},{"alias_kind":"pith_short_16","alias_value":"LPH6T5NM6EZ3BVPT","created_at":"2026-05-28T01:05:12.457524+00:00"},{"alias_kind":"pith_short_8","alias_value":"LPH6T5NM","created_at":"2026-05-28T01:05:12.457524+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LPH6T5NM6EZ3BVPTSI7NPVSVZ6","json":"https://pith.science/pith/LPH6T5NM6EZ3BVPTSI7NPVSVZ6.json","graph_json":"https://pith.science/api/pith-number/LPH6T5NM6EZ3BVPTSI7NPVSVZ6/graph.json","events_json":"https://pith.science/api/pith-number/LPH6T5NM6EZ3BVPTSI7NPVSVZ6/events.json","paper":"https://pith.science/paper/LPH6T5NM"},"agent_actions":{"view_html":"https://pith.science/pith/LPH6T5NM6EZ3BVPTSI7NPVSVZ6","download_json":"https://pith.science/pith/LPH6T5NM6EZ3BVPTSI7NPVSVZ6.json","view_paper":"https://pith.science/paper/LPH6T5NM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.01735&json=true","fetch_graph":"https://pith.science/api/pith-number/LPH6T5NM6EZ3BVPTSI7NPVSVZ6/graph.json","fetch_events":"https://pith.science/api/pith-number/LPH6T5NM6EZ3BVPTSI7NPVSVZ6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LPH6T5NM6EZ3BVPTSI7NPVSVZ6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LPH6T5NM6EZ3BVPTSI7NPVSVZ6/action/storage_attestation","attest_author":"https://pith.science/pith/LPH6T5NM6EZ3BVPTSI7NPVSVZ6/action/author_attestation","sign_citation":"https://pith.science/pith/LPH6T5NM6EZ3BVPTSI7NPVSVZ6/action/citation_signature","submit_replication":"https://pith.science/pith/LPH6T5NM6EZ3BVPTSI7NPVSVZ6/action/replication_record"}},"created_at":"2026-05-28T01:05:12.457524+00:00","updated_at":"2026-05-28T01:05:12.457524+00:00"}