{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EUYV7QXB4K6AGJN6DT5RJ5OUIT","short_pith_number":"pith:EUYV7QXB","schema_version":"1.0","canonical_sha256":"25315fc2e1e2bc0325be1cfb14f5d444edba6af4d9bbff92d0478ae8fa41d892","source":{"kind":"arxiv","id":"2505.16008","version":1},"attestation_state":"computed","paper":{"title":"LAGO: Few-shot Crosslingual Embedding Inversion Attacks via Language Similarity-Aware Graph Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.CL","authors_text":"Johannes Bjerva, Qiongxiu Li, Sokol Kosta, Wenrui Yu, Yiyi Chen","submitted_at":"2025-05-21T20:48:24Z","abstract_excerpt":"We propose LAGO - Language Similarity-Aware Graph Optimization - a novel approach for few-shot cross-lingual embedding inversion attacks, addressing critical privacy vulnerabilities in multilingual NLP systems. Unlike prior work in embedding inversion attacks that treat languages independently, LAGO explicitly models linguistic relationships through a graph-based constrained distributed optimization framework. By integrating syntactic and lexical similarity as edge constraints, our method enables collaborative parameter learning across related languages. Theoretically, we show this formulation"},"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":"2505.16008","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-21T20:48:24Z","cross_cats_sorted":["cs.AI","cs.CR"],"title_canon_sha256":"74e812dce97480143ee2cae784a5db773b58a9a7f3e89983d9ba9c65ff913c97","abstract_canon_sha256":"05e6c00b9d3f94dc7546f0c6911b4b7428f5aff543d61306763b3aae82b5c803"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:07:28.517854Z","signature_b64":"lbajSQ79HOYARfIXb8zVHq7krSKQWEtXqRANRbdmvMX/mkz54S+EAtSdiv20qwZ2vMATfSkyDorjTj/9DlLQAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"25315fc2e1e2bc0325be1cfb14f5d444edba6af4d9bbff92d0478ae8fa41d892","last_reissued_at":"2026-07-05T11:07:28.517275Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:07:28.517275Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LAGO: Few-shot Crosslingual Embedding Inversion Attacks via Language Similarity-Aware Graph Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.CL","authors_text":"Johannes Bjerva, Qiongxiu Li, Sokol Kosta, Wenrui Yu, Yiyi Chen","submitted_at":"2025-05-21T20:48:24Z","abstract_excerpt":"We propose LAGO - Language Similarity-Aware Graph Optimization - a novel approach for few-shot cross-lingual embedding inversion attacks, addressing critical privacy vulnerabilities in multilingual NLP systems. Unlike prior work in embedding inversion attacks that treat languages independently, LAGO explicitly models linguistic relationships through a graph-based constrained distributed optimization framework. By integrating syntactic and lexical similarity as edge constraints, our method enables collaborative parameter learning across related languages. Theoretically, we show this formulation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.16008","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/2505.16008/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":"2505.16008","created_at":"2026-07-05T11:07:28.517341+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.16008v1","created_at":"2026-07-05T11:07:28.517341+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.16008","created_at":"2026-07-05T11:07:28.517341+00:00"},{"alias_kind":"pith_short_12","alias_value":"EUYV7QXB4K6A","created_at":"2026-07-05T11:07:28.517341+00:00"},{"alias_kind":"pith_short_16","alias_value":"EUYV7QXB4K6AGJN6","created_at":"2026-07-05T11:07:28.517341+00:00"},{"alias_kind":"pith_short_8","alias_value":"EUYV7QXB","created_at":"2026-07-05T11:07:28.517341+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27976","citing_title":"SHARD: cell-keyed residual splitting for alignment-resistant private dense retrieval","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27976","citing_title":"SHARD: cell-keyed residual splitting for alignment-resistant private dense retrieval","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EUYV7QXB4K6AGJN6DT5RJ5OUIT","json":"https://pith.science/pith/EUYV7QXB4K6AGJN6DT5RJ5OUIT.json","graph_json":"https://pith.science/api/pith-number/EUYV7QXB4K6AGJN6DT5RJ5OUIT/graph.json","events_json":"https://pith.science/api/pith-number/EUYV7QXB4K6AGJN6DT5RJ5OUIT/events.json","paper":"https://pith.science/paper/EUYV7QXB"},"agent_actions":{"view_html":"https://pith.science/pith/EUYV7QXB4K6AGJN6DT5RJ5OUIT","download_json":"https://pith.science/pith/EUYV7QXB4K6AGJN6DT5RJ5OUIT.json","view_paper":"https://pith.science/paper/EUYV7QXB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.16008&json=true","fetch_graph":"https://pith.science/api/pith-number/EUYV7QXB4K6AGJN6DT5RJ5OUIT/graph.json","fetch_events":"https://pith.science/api/pith-number/EUYV7QXB4K6AGJN6DT5RJ5OUIT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EUYV7QXB4K6AGJN6DT5RJ5OUIT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EUYV7QXB4K6AGJN6DT5RJ5OUIT/action/storage_attestation","attest_author":"https://pith.science/pith/EUYV7QXB4K6AGJN6DT5RJ5OUIT/action/author_attestation","sign_citation":"https://pith.science/pith/EUYV7QXB4K6AGJN6DT5RJ5OUIT/action/citation_signature","submit_replication":"https://pith.science/pith/EUYV7QXB4K6AGJN6DT5RJ5OUIT/action/replication_record"}},"created_at":"2026-07-05T11:07:28.517341+00:00","updated_at":"2026-07-05T11:07:28.517341+00:00"}