{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:3A3NSDLDDAURSZEKB2WUDWSEKN","short_pith_number":"pith:3A3NSDLD","schema_version":"1.0","canonical_sha256":"d836d90d63182919648a0ead41da445340ce937cfd128a535c945fe435ff69ef","source":{"kind":"arxiv","id":"2605.13538","version":1},"attestation_state":"computed","paper":{"title":"Locale-Conditioned Few-Shot Prompting Mitigates Demonstration Regurgitation in On-Device PII Substitution with Small Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Locale-conditioned rotating few-shot prompts stop small language models from echoing demonstration examples during on-device PII substitution.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Anuj Sadani, Deepak Kumar","submitted_at":"2026-05-13T13:47:11Z","abstract_excerpt":"Personally Identifiable Information (PII) redaction usually replaces detected entities with placeholder tokens such as [PERSON], destroying the downstream utility of the redacted text for retrieval and Named Entity Recognition (NER) training. We propose a fully on-device pipeline that substitutes PII with consistent, type-preserving fake values: a 1.5 B mixture-of-experts token classifier (openai/privacy-filter) detects spans, a 1-bit Bonsai-1.7B Small Language Model (SLM) proposes contextual surrogates for names, addresses, and dates, and a rule-based generator (faker) handles patterned field"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.13538","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-13T13:47:11Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"71ca1095213f9418fa258a120300c4beee4c761bb6f28c94731d6b266582527e","abstract_canon_sha256":"08982f296d3b7011c8b7b2fdf6bec4c5c580d305c1cc7386049148fae213f8dc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:44:24.089214Z","signature_b64":"UiiGXX66GqPyk1m3keJIB02P+V1lkUMgi/TVIISJ75wsBSzxgqypbEdhHkYG6O7UB9AclxzmKuuQDANrTUErDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d836d90d63182919648a0ead41da445340ce937cfd128a535c945fe435ff69ef","last_reissued_at":"2026-05-18T02:44:24.088768Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:44:24.088768Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Locale-Conditioned Few-Shot Prompting Mitigates Demonstration Regurgitation in On-Device PII Substitution with Small Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Locale-conditioned rotating few-shot prompts stop small language models from echoing demonstration examples during on-device PII substitution.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Anuj Sadani, Deepak Kumar","submitted_at":"2026-05-13T13:47:11Z","abstract_excerpt":"Personally Identifiable Information (PII) redaction usually replaces detected entities with placeholder tokens such as [PERSON], destroying the downstream utility of the redacted text for retrieval and Named Entity Recognition (NER) training. We propose a fully on-device pipeline that substitutes PII with consistent, type-preserving fake values: a 1.5 B mixture-of-experts token classifier (openai/privacy-filter) detects spans, a 1-bit Bonsai-1.7B Small Language Model (SLM) proposes contextual surrogates for names, addresses, and dates, and a rule-based generator (faker) handles patterned field"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"With the fix, 482/482 unique Bonsai-1.7B calls succeed (no echoes) and produce locale-correct surrogates, although the SLM still copies from a small same-locale demonstration pool - a residual narrowness we quantify.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the observed NER performance gap is caused primarily by reduced variety in SLM outputs rather than by other unmeasured differences in the 160/40 subset or by the choice of XGLM-564M as the multilingual evaluator.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Locale-conditioned rotating few-shot prompting eliminates demonstration regurgitation in 1.7B SLMs for PII substitution while producing more natural text than rule-based methods, though downstream NER training benefits more from synthetic variety than naturalness.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Locale-conditioned rotating few-shot prompts stop small language models from echoing demonstration examples during on-device PII substitution.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"1767d0d1798e97fd46bb4e9176590148e9eb44ae983ad5d85fb458dc47ca6388"},"source":{"id":"2605.13538","kind":"arxiv","version":1},"verdict":{"id":"fddab2d7-c488-4d6a-8201-d6bc53eabbb7","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T19:00:57.673618Z","strongest_claim":"With the fix, 482/482 unique Bonsai-1.7B calls succeed (no echoes) and produce locale-correct surrogates, although the SLM still copies from a small same-locale demonstration pool - a residual narrowness we quantify.","one_line_summary":"Locale-conditioned rotating few-shot prompting eliminates demonstration regurgitation in 1.7B SLMs for PII substitution while producing more natural text than rule-based methods, though downstream NER training benefits more from synthetic variety than naturalness.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the observed NER performance gap is caused primarily by reduced variety in SLM outputs rather than by other unmeasured differences in the 160/40 subset or by the choice of XGLM-564M as the multilingual evaluator.","pith_extraction_headline":"Locale-conditioned rotating few-shot prompts stop small language models from echoing demonstration examples during on-device PII substitution."},"references":{"count":16,"sample":[{"doi":"","year":2024,"title":"Faker: a Python package that generates fake data","work_id":"b1918f04-be18-4cdc-8b1b-8098eac3c9ec","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"llama.cpp: Port of LLaMA models in C/C++","work_id":"6b334f6b-fac2-4e0c-910e-43ed4197b5c2","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"spaCy: Industrial-strength natural language processing in Python","work_id":"8878703d-2330-47d2-81ec-6f3b4c7eacb6","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Few-shot learning with multilingual language models","work_id":"32d56185-d394-41f6-a038-08c943a4650c","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Fantastically ordered prompts and where to find them: Overcoming few-shot prompt order sensitivity","work_id":"c6ecb82b-59ca-437f-91d8-470531f9f7c8","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":16,"snapshot_sha256":"8f7fc02398e1eadab2ea9d9fadc8e255a1580afbb705951483156df5e9d4b78b","internal_anchors":3},"formal_canon":{"evidence_count":1,"snapshot_sha256":"13cea1224d6f07fff5395eb607961feb0d0eb9b58ef3be5215275e4827c88642"},"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":"2605.13538","created_at":"2026-05-18T02:44:24.088840+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.13538v1","created_at":"2026-05-18T02:44:24.088840+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13538","created_at":"2026-05-18T02:44:24.088840+00:00"},{"alias_kind":"pith_short_12","alias_value":"3A3NSDLDDAUR","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"3A3NSDLDDAURSZEK","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"3A3NSDLD","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":1,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3A3NSDLDDAURSZEKB2WUDWSEKN","json":"https://pith.science/pith/3A3NSDLDDAURSZEKB2WUDWSEKN.json","graph_json":"https://pith.science/api/pith-number/3A3NSDLDDAURSZEKB2WUDWSEKN/graph.json","events_json":"https://pith.science/api/pith-number/3A3NSDLDDAURSZEKB2WUDWSEKN/events.json","paper":"https://pith.science/paper/3A3NSDLD"},"agent_actions":{"view_html":"https://pith.science/pith/3A3NSDLDDAURSZEKB2WUDWSEKN","download_json":"https://pith.science/pith/3A3NSDLDDAURSZEKB2WUDWSEKN.json","view_paper":"https://pith.science/paper/3A3NSDLD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.13538&json=true","fetch_graph":"https://pith.science/api/pith-number/3A3NSDLDDAURSZEKB2WUDWSEKN/graph.json","fetch_events":"https://pith.science/api/pith-number/3A3NSDLDDAURSZEKB2WUDWSEKN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3A3NSDLDDAURSZEKB2WUDWSEKN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3A3NSDLDDAURSZEKB2WUDWSEKN/action/storage_attestation","attest_author":"https://pith.science/pith/3A3NSDLDDAURSZEKB2WUDWSEKN/action/author_attestation","sign_citation":"https://pith.science/pith/3A3NSDLDDAURSZEKB2WUDWSEKN/action/citation_signature","submit_replication":"https://pith.science/pith/3A3NSDLDDAURSZEKB2WUDWSEKN/action/replication_record"}},"created_at":"2026-05-18T02:44:24.088840+00:00","updated_at":"2026-05-18T02:44:24.088840+00:00"}