{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DYYTTHVJF7K5IEJHWZMZFLLGP5","short_pith_number":"pith:DYYTTHVJ","schema_version":"1.0","canonical_sha256":"1e31399ea92fd5d41127b65992ad667f505cafce739752e632f0cdd5e45435c6","source":{"kind":"arxiv","id":"2603.23231","version":2},"attestation_state":"computed","paper":{"title":"PERMA: Benchmarking Personalized Memory Agents via Event-Driven Preference and Realistic Task Environments","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bo Tang, Chao Zhang, Derong Xu, Enhong Chen, Feiyu Xiong, Haotian Zhang, Jia Li, Junda Lin, Junyi Zhu, Long Shu, Shuochen Liu, Tong Xu, Yuhao Chen, Zhiyu Li","submitted_at":"2026-03-24T14:04:11Z","abstract_excerpt":"Empowering large language models with long-term memory is crucial for building agents that adapt to users' evolving needs. Existing evaluations of this capability typically interleave preference-related dialogues with irrelevant conversations, reducing the task to needle-in-a-haystack retrieval while ignoring relationships between events driving user preference evolution. Such settings overlook a fundamental characteristic of real-world personalization: preferences emerge gradually and accumulate across interactions within noisy contexts. To bridge this gap, we introduce PERMA, a benchmark des"},"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":"2603.23231","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-03-24T14:04:11Z","cross_cats_sorted":[],"title_canon_sha256":"fd67d36b77b681afee8e91d804482181879ccaebda4862d08bb5a77a179c93e4","abstract_canon_sha256":"705beb731dcd3307383785d8afef8b410cb52d76b9b2b5e46ac714b273947968"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:29.855539Z","signature_b64":"nCLS1aGp9dD27fDL/V5nQ6OzaNbiE0pAY0pBFT5ykdZobQBIIMa3008zxtOm9D67zndFQ0XWzuYcHgfHVUvZCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1e31399ea92fd5d41127b65992ad667f505cafce739752e632f0cdd5e45435c6","last_reissued_at":"2026-05-20T00:04:29.854680Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:29.854680Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PERMA: Benchmarking Personalized Memory Agents via Event-Driven Preference and Realistic Task Environments","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bo Tang, Chao Zhang, Derong Xu, Enhong Chen, Feiyu Xiong, Haotian Zhang, Jia Li, Junda Lin, Junyi Zhu, Long Shu, Shuochen Liu, Tong Xu, Yuhao Chen, Zhiyu Li","submitted_at":"2026-03-24T14:04:11Z","abstract_excerpt":"Empowering large language models with long-term memory is crucial for building agents that adapt to users' evolving needs. Existing evaluations of this capability typically interleave preference-related dialogues with irrelevant conversations, reducing the task to needle-in-a-haystack retrieval while ignoring relationships between events driving user preference evolution. Such settings overlook a fundamental characteristic of real-world personalization: preferences emerge gradually and accumulate across interactions within noisy contexts. To bridge this gap, we introduce PERMA, a benchmark des"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.23231","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/2603.23231/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":"2603.23231","created_at":"2026-05-20T00:04:29.854841+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.23231v2","created_at":"2026-05-20T00:04:29.854841+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.23231","created_at":"2026-05-20T00:04:29.854841+00:00"},{"alias_kind":"pith_short_12","alias_value":"DYYTTHVJF7K5","created_at":"2026-05-20T00:04:29.854841+00:00"},{"alias_kind":"pith_short_16","alias_value":"DYYTTHVJF7K5IEJH","created_at":"2026-05-20T00:04:29.854841+00:00"},{"alias_kind":"pith_short_8","alias_value":"DYYTTHVJ","created_at":"2026-05-20T00:04:29.854841+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2605.14678","citing_title":"$\\pi$-Bench: Evaluating Proactive Personal Assistant Agents in Long-Horizon Workflows","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14678","citing_title":"$\\pi$-Bench: Evaluating Proactive Personal Assistant Agents in Long-Horizon Workflows","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2604.20261","citing_title":"Memory-Augmented LLM-based Multi-Agent System for Automated Feature Generation on Tabular Data","ref_index":74,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DYYTTHVJF7K5IEJHWZMZFLLGP5","json":"https://pith.science/pith/DYYTTHVJF7K5IEJHWZMZFLLGP5.json","graph_json":"https://pith.science/api/pith-number/DYYTTHVJF7K5IEJHWZMZFLLGP5/graph.json","events_json":"https://pith.science/api/pith-number/DYYTTHVJF7K5IEJHWZMZFLLGP5/events.json","paper":"https://pith.science/paper/DYYTTHVJ"},"agent_actions":{"view_html":"https://pith.science/pith/DYYTTHVJF7K5IEJHWZMZFLLGP5","download_json":"https://pith.science/pith/DYYTTHVJF7K5IEJHWZMZFLLGP5.json","view_paper":"https://pith.science/paper/DYYTTHVJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.23231&json=true","fetch_graph":"https://pith.science/api/pith-number/DYYTTHVJF7K5IEJHWZMZFLLGP5/graph.json","fetch_events":"https://pith.science/api/pith-number/DYYTTHVJF7K5IEJHWZMZFLLGP5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DYYTTHVJF7K5IEJHWZMZFLLGP5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DYYTTHVJF7K5IEJHWZMZFLLGP5/action/storage_attestation","attest_author":"https://pith.science/pith/DYYTTHVJF7K5IEJHWZMZFLLGP5/action/author_attestation","sign_citation":"https://pith.science/pith/DYYTTHVJF7K5IEJHWZMZFLLGP5/action/citation_signature","submit_replication":"https://pith.science/pith/DYYTTHVJF7K5IEJHWZMZFLLGP5/action/replication_record"}},"created_at":"2026-05-20T00:04:29.854841+00:00","updated_at":"2026-05-20T00:04:29.854841+00:00"}