{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:4AZSR7K745O6Z5UJGIBSVAVXAN","short_pith_number":"pith:4AZSR7K7","schema_version":"1.0","canonical_sha256":"e03328fd5fe75decf68932032a82b703623a73a258bfd48821f2d72df7f4bb4a","source":{"kind":"arxiv","id":"2501.10915","version":1},"attestation_state":"computed","paper":{"title":"LegalGuardian: A Privacy-Preserving Framework for Secure Integration of Large Language Models in Legal Practice","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CR","cs.IR"],"primary_cat":"cs.CL","authors_text":"Hakan T. Otal, M. Abdullah Canbaz, M. Mikail Demir","submitted_at":"2025-01-19T01:43:42Z","abstract_excerpt":"Large Language Models (LLMs) hold promise for advancing legal practice by automating complex tasks and improving access to justice. However, their adoption is limited by concerns over client confidentiality, especially when lawyers include sensitive Personally Identifiable Information (PII) in prompts, risking unauthorized data exposure. To mitigate this, we introduce LegalGuardian, a lightweight, privacy-preserving framework tailored for lawyers using LLM-based tools. LegalGuardian employs Named Entity Recognition (NER) techniques and local LLMs to mask and unmask confidential PII within prom"},"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":"2501.10915","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-19T01:43:42Z","cross_cats_sorted":["cs.CR","cs.IR"],"title_canon_sha256":"50fd43c23def8a439b764d6ed32fa6424360cca292798f7356e7ea17de4fd2e9","abstract_canon_sha256":"176ef12a949e22b51bc821de21ec0805c52ca9d35ff074928e25a2da8ee45d26"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:03:00.700226Z","signature_b64":"feYICHDVfSiB5VpC3W4U489sEYO7b+iahm4zEkSb51mcwmGODJMwMJ0ZGNHNvDJlbl4aiuetToEfFadEpnOAAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e03328fd5fe75decf68932032a82b703623a73a258bfd48821f2d72df7f4bb4a","last_reissued_at":"2026-07-05T10:03:00.699704Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:03:00.699704Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LegalGuardian: A Privacy-Preserving Framework for Secure Integration of Large Language Models in Legal Practice","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CR","cs.IR"],"primary_cat":"cs.CL","authors_text":"Hakan T. Otal, M. Abdullah Canbaz, M. Mikail Demir","submitted_at":"2025-01-19T01:43:42Z","abstract_excerpt":"Large Language Models (LLMs) hold promise for advancing legal practice by automating complex tasks and improving access to justice. However, their adoption is limited by concerns over client confidentiality, especially when lawyers include sensitive Personally Identifiable Information (PII) in prompts, risking unauthorized data exposure. To mitigate this, we introduce LegalGuardian, a lightweight, privacy-preserving framework tailored for lawyers using LLM-based tools. LegalGuardian employs Named Entity Recognition (NER) techniques and local LLMs to mask and unmask confidential PII within prom"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.10915","kind":"arxiv","version":1},"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/2501.10915/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":"2501.10915","created_at":"2026-07-05T10:03:00.699772+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.10915v1","created_at":"2026-07-05T10:03:00.699772+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.10915","created_at":"2026-07-05T10:03:00.699772+00:00"},{"alias_kind":"pith_short_12","alias_value":"4AZSR7K745O6","created_at":"2026-07-05T10:03:00.699772+00:00"},{"alias_kind":"pith_short_16","alias_value":"4AZSR7K745O6Z5UJ","created_at":"2026-07-05T10:03:00.699772+00:00"},{"alias_kind":"pith_short_8","alias_value":"4AZSR7K7","created_at":"2026-07-05T10:03:00.699772+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2505.14549","citing_title":"Can Large Language Models Really Recognize Your Name?","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17691","citing_title":"Validate Your Authority: Benchmarking LLMs on Multi-Label Precedent Treatment Classification","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4AZSR7K745O6Z5UJGIBSVAVXAN","json":"https://pith.science/pith/4AZSR7K745O6Z5UJGIBSVAVXAN.json","graph_json":"https://pith.science/api/pith-number/4AZSR7K745O6Z5UJGIBSVAVXAN/graph.json","events_json":"https://pith.science/api/pith-number/4AZSR7K745O6Z5UJGIBSVAVXAN/events.json","paper":"https://pith.science/paper/4AZSR7K7"},"agent_actions":{"view_html":"https://pith.science/pith/4AZSR7K745O6Z5UJGIBSVAVXAN","download_json":"https://pith.science/pith/4AZSR7K745O6Z5UJGIBSVAVXAN.json","view_paper":"https://pith.science/paper/4AZSR7K7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.10915&json=true","fetch_graph":"https://pith.science/api/pith-number/4AZSR7K745O6Z5UJGIBSVAVXAN/graph.json","fetch_events":"https://pith.science/api/pith-number/4AZSR7K745O6Z5UJGIBSVAVXAN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4AZSR7K745O6Z5UJGIBSVAVXAN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4AZSR7K745O6Z5UJGIBSVAVXAN/action/storage_attestation","attest_author":"https://pith.science/pith/4AZSR7K745O6Z5UJGIBSVAVXAN/action/author_attestation","sign_citation":"https://pith.science/pith/4AZSR7K745O6Z5UJGIBSVAVXAN/action/citation_signature","submit_replication":"https://pith.science/pith/4AZSR7K745O6Z5UJGIBSVAVXAN/action/replication_record"}},"created_at":"2026-07-05T10:03:00.699772+00:00","updated_at":"2026-07-05T10:03:00.699772+00:00"}