{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NU2VFINOYVEREFO6MSXY22IRHY","merge_version":"pith-open-graph-merge-v1","event_count":3,"valid_event_count":3,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"7cc410bd56c9ef7568e17b1c969c02388d3de92d03d89ba0911a65cfc04f7c51","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-20T23:14:36Z","title_canon_sha256":"262856f87c765885a3614c7706585cb94be385f3cb214219a64eb9b74d1c2a31"},"schema_version":"1.0","source":{"id":"2607.21635","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.21635","created_at":"2026-07-27T00:20:12Z"},{"alias_kind":"arxiv_version","alias_value":"2607.21635v1","created_at":"2026-07-27T00:20:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.21635","created_at":"2026-07-27T00:20:12Z"},{"alias_kind":"pith_short_12","alias_value":"NU2VFINOYVER","created_at":"2026-07-27T00:20:12Z"},{"alias_kind":"pith_short_16","alias_value":"NU2VFINOYVEREFO6","created_at":"2026-07-27T00:20:12Z"},{"alias_kind":"pith_short_8","alias_value":"NU2VFINO","created_at":"2026-07-27T00:20:12Z"}],"graph_snapshots":[{"event_id":"sha256:0ea184a2b1c247f9e3de3645a03264d07b4ce2ab01af23400a4f18154ce45cce","target":"graph","created_at":"2026-07-27T00:20:12Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2607.21635/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Personal agents maintain memories, learned skills, tool configurations, and policy state that evolve with each user. Existing agent benchmarks often evaluate these capabilities in isolation: tool benchmarks test invocation under fixed APIs, memory benchmarks test recall or forgetting, and safety benchmarks test static policy compliance. We argue that personal-agent evaluation requires a different protocol: replaying the same temporal intervention across different persistent user-conditioned states and measuring how failures propagate across agent components. We formalize this requirement as fo","authors_text":"Junxian You, Pin Qian, Qiaolin Yu, Su Wang, Xiaoyuan Wang, Yihang Chen, Zhicheng Wang, Zhitong Guo","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-20T23:14:36Z","title":"Toward User-Conditioned Evaluation of Personal LLM Agents under Temporal Interventions"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.21635","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:9623ff18c21e847db6868b42e331c467ce073cba3b80bd4e070fae5f6df7ddcf","target":"record","created_at":"2026-07-27T00:20:12Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"7cc410bd56c9ef7568e17b1c969c02388d3de92d03d89ba0911a65cfc04f7c51","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-20T23:14:36Z","title_canon_sha256":"262856f87c765885a3614c7706585cb94be385f3cb214219a64eb9b74d1c2a31"},"schema_version":"1.0","source":{"id":"2607.21635","kind":"arxiv","version":1}},"canonical_sha256":"6d3552a1aec5491215de64af8d69113e1a6e00c0ebca4b4a8c47e445e3dcd6cb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6d3552a1aec5491215de64af8d69113e1a6e00c0ebca4b4a8c47e445e3dcd6cb","first_computed_at":"2026-07-27T00:20:12.215557Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-27T00:20:12.215557Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HcqX0B3eqH9/fLLqrsVruM3dW5nMaFpTHv1t3PcFldWpplsjOVUGtBY06WcHMWiqrWlBF90igAV3bhZbKOVoAg==","signature_status":"signed_v1","signed_at":"2026-07-27T00:20:12.216446Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.21635","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9623ff18c21e847db6868b42e331c467ce073cba3b80bd4e070fae5f6df7ddcf","sha256:0ea184a2b1c247f9e3de3645a03264d07b4ce2ab01af23400a4f18154ce45cce","sha256:634ba29e426d8afeb145a8153d3b396152f3c8314a5f4cbddcc3c201d4c98425"],"state_sha256":"ae8c5c5cdcfc6b4a8248914b908750be50ff76dad4bc6f1060ed2ac5db11144d"}