{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:SVJR56AFLUSQCWA6Q3K4ME2NJI","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"45ae21df8e42a989e73c1a490f57b134378476e18039e193ff5182ce924f8eec","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.HC","submitted_at":"2026-04-07T19:42:36Z","title_canon_sha256":"fb55cb039f9369043961e9dee1ee8f86164eed8387882caf140540c8842094d0"},"schema_version":"1.0","source":{"id":"2605.20204","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.20204","created_at":"2026-05-21T00:04:21Z"},{"alias_kind":"arxiv_version","alias_value":"2605.20204v1","created_at":"2026-05-21T00:04:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.20204","created_at":"2026-05-21T00:04:21Z"},{"alias_kind":"pith_short_12","alias_value":"SVJR56AFLUSQ","created_at":"2026-05-21T00:04:21Z"},{"alias_kind":"pith_short_16","alias_value":"SVJR56AFLUSQCWA6","created_at":"2026-05-21T00:04:21Z"},{"alias_kind":"pith_short_8","alias_value":"SVJR56AF","created_at":"2026-05-21T00:04:21Z"}],"graph_snapshots":[{"event_id":"sha256:54da4db4dadb594f1e93f08bd6a9a8ee48c7036c77f36e6fdee285fe41f37173","target":"graph","created_at":"2026-05-21T00:04:21Z","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/2605.20204/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLM-based user simulation is the primary mechanism for end-to-end agent evaluation, yet simulated users are poor proxies for real humans: unconstrained LLM defaults produce a Formalism Ceiling (style match rates of 6-8% against real users), while hand-crafted behavioral directives trigger Directive Amplification, where models hyper-interpret instructions into unnatural behavioral extremes that vary dramatically across simulator models. We present RealUserSim, the first user simulation framework grounded in real behavioral data. From 14,000+ authentic human-LLM conversations (WildChat), we extr","authors_text":"Huan Wang, Jielin Qiu, Juntao Tan, Liangwei Yang, Ming Zhu, Rithesh Murthy, Shelby Heinecke, Silvio Savarese, Wenting Zhao","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.HC","submitted_at":"2026-04-07T19:42:36Z","title":"RealUserSim: Bridging the Reality Gap in Agent Benchmarking via Grounded User Simulation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.20204","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":"sha256:12c1a1e64faa8a9da70a139b4e55f8aac542dead4ca96029c7fc60a77b81d1ca","target":"record","created_at":"2026-05-21T00:04:21Z","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":"45ae21df8e42a989e73c1a490f57b134378476e18039e193ff5182ce924f8eec","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.HC","submitted_at":"2026-04-07T19:42:36Z","title_canon_sha256":"fb55cb039f9369043961e9dee1ee8f86164eed8387882caf140540c8842094d0"},"schema_version":"1.0","source":{"id":"2605.20204","kind":"arxiv","version":1}},"canonical_sha256":"95531ef8055d2501581e86d5c6134d4a3796d5bf9260fb5231af35a05bc9d9e6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"95531ef8055d2501581e86d5c6134d4a3796d5bf9260fb5231af35a05bc9d9e6","first_computed_at":"2026-05-21T00:04:21.003815Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-21T00:04:21.003815Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pmzox9WLuIz0VWgc+US8CEbZ9STSFNy8GzJRV6VRe7yyzgi01wBHHxcqzQHuQynQQOJLcCF1UAabZRxbBitGAA==","signature_status":"signed_v1","signed_at":"2026-05-21T00:04:21.004585Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.20204","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:12c1a1e64faa8a9da70a139b4e55f8aac542dead4ca96029c7fc60a77b81d1ca","sha256:54da4db4dadb594f1e93f08bd6a9a8ee48c7036c77f36e6fdee285fe41f37173"],"state_sha256":"1574aa49a7336bebc2061d6ba9cf5761d641079b4e0512ceaadf4da03cb6bee0"}