{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HNS63G2YXYAGE5JTG3PLFOYQEJ","short_pith_number":"pith:HNS63G2Y","schema_version":"1.0","canonical_sha256":"3b65ed9b58be0062753336deb2bb1022415fe46f8822027cef910b447a80473d","source":{"kind":"arxiv","id":"2502.20616","version":2},"attestation_state":"computed","paper":{"title":"PersonaBench: Evaluating AI Models on Understanding Personal Information through Accessing (Synthetic) Private User Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Caiming Xiong, Huan Wang, Jianguo Zhang, Juntao Tan, Liangwei Yang, Ming Zhu, Rithesh Murthy, Shelby Heinecke, Shirley Kokane, Silvio Savarese, Tulika Manoj Awalgaonkar, Weiran Yao, Zhiwei Liu, Zuxin Liu","submitted_at":"2025-02-28T00:43:35Z","abstract_excerpt":"Personalization is critical in AI assistants, particularly in the context of private AI models that work with individual users. A key scenario in this domain involves enabling AI models to access and interpret a user's private data (e.g., conversation history, user-AI interactions, app usage) to understand personal details such as biographical information, preferences, and social connections. However, due to the sensitive nature of such data, there are no publicly available datasets that allow us to assess an AI model's ability to understand users through direct access to personal information."},"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":"2502.20616","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-02-28T00:43:35Z","cross_cats_sorted":[],"title_canon_sha256":"4ae00a42d6d7409d1cdda25e10d1047439692215d17433f5b32450de633dc4b4","abstract_canon_sha256":"4c309b7e7dd44cd09d96619cb8c3f2664b33a29cb2c6dedc35e6bb372e921d03"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:56:53.372594Z","signature_b64":"8FvtahmN47XhYR2FoISItLosqq14IXQSDI5e98zAK5bsb1wnboV+ZJQAAgeSbY9M/HCAKhcKm4CdwJqj9NjDDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3b65ed9b58be0062753336deb2bb1022415fe46f8822027cef910b447a80473d","last_reissued_at":"2026-07-05T11:56:53.372059Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:56:53.372059Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PersonaBench: Evaluating AI Models on Understanding Personal Information through Accessing (Synthetic) Private User Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Caiming Xiong, Huan Wang, Jianguo Zhang, Juntao Tan, Liangwei Yang, Ming Zhu, Rithesh Murthy, Shelby Heinecke, Shirley Kokane, Silvio Savarese, Tulika Manoj Awalgaonkar, Weiran Yao, Zhiwei Liu, Zuxin Liu","submitted_at":"2025-02-28T00:43:35Z","abstract_excerpt":"Personalization is critical in AI assistants, particularly in the context of private AI models that work with individual users. A key scenario in this domain involves enabling AI models to access and interpret a user's private data (e.g., conversation history, user-AI interactions, app usage) to understand personal details such as biographical information, preferences, and social connections. However, due to the sensitive nature of such data, there are no publicly available datasets that allow us to assess an AI model's ability to understand users through direct access to personal information."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20616","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/2502.20616/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":"2502.20616","created_at":"2026-07-05T11:56:53.372122+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.20616v2","created_at":"2026-07-05T11:56:53.372122+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20616","created_at":"2026-07-05T11:56:53.372122+00:00"},{"alias_kind":"pith_short_12","alias_value":"HNS63G2YXYAG","created_at":"2026-07-05T11:56:53.372122+00:00"},{"alias_kind":"pith_short_16","alias_value":"HNS63G2YXYAGE5JT","created_at":"2026-07-05T11:56:53.372122+00:00"},{"alias_kind":"pith_short_8","alias_value":"HNS63G2Y","created_at":"2026-07-05T11:56:53.372122+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26403","citing_title":"ProfileFoundry: A Synthetic Person-Object Substrate for Privacy, Memory, and Tool-Use Evaluation in LLM Agent","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2603.23231","citing_title":"PERMA: Benchmarking Personalized Memory Agents via Event-Driven Preference and Realistic Task Environments","ref_index":49,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HNS63G2YXYAGE5JTG3PLFOYQEJ","json":"https://pith.science/pith/HNS63G2YXYAGE5JTG3PLFOYQEJ.json","graph_json":"https://pith.science/api/pith-number/HNS63G2YXYAGE5JTG3PLFOYQEJ/graph.json","events_json":"https://pith.science/api/pith-number/HNS63G2YXYAGE5JTG3PLFOYQEJ/events.json","paper":"https://pith.science/paper/HNS63G2Y"},"agent_actions":{"view_html":"https://pith.science/pith/HNS63G2YXYAGE5JTG3PLFOYQEJ","download_json":"https://pith.science/pith/HNS63G2YXYAGE5JTG3PLFOYQEJ.json","view_paper":"https://pith.science/paper/HNS63G2Y","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.20616&json=true","fetch_graph":"https://pith.science/api/pith-number/HNS63G2YXYAGE5JTG3PLFOYQEJ/graph.json","fetch_events":"https://pith.science/api/pith-number/HNS63G2YXYAGE5JTG3PLFOYQEJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HNS63G2YXYAGE5JTG3PLFOYQEJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HNS63G2YXYAGE5JTG3PLFOYQEJ/action/storage_attestation","attest_author":"https://pith.science/pith/HNS63G2YXYAGE5JTG3PLFOYQEJ/action/author_attestation","sign_citation":"https://pith.science/pith/HNS63G2YXYAGE5JTG3PLFOYQEJ/action/citation_signature","submit_replication":"https://pith.science/pith/HNS63G2YXYAGE5JTG3PLFOYQEJ/action/replication_record"}},"created_at":"2026-07-05T11:56:53.372122+00:00","updated_at":"2026-07-05T11:56:53.372122+00:00"}