{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:25LINL6LOXBRFLW5JRWIAMAVDX","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":"3886f78d77a3d4b52228b02c1cbcdc4a75c50e7ba5434757f6896e9c5814de22","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-09T16:39:32Z","title_canon_sha256":"54cc92bbceba1bccbf0c6b5262dd0c63e374ab5bfc4d27989aebabb8717cd9c8"},"schema_version":"1.0","source":{"id":"2606.11079","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.11079","created_at":"2026-06-10T01:11:08Z"},{"alias_kind":"arxiv_version","alias_value":"2606.11079v1","created_at":"2026-06-10T01:11:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.11079","created_at":"2026-06-10T01:11:08Z"},{"alias_kind":"pith_short_12","alias_value":"25LINL6LOXBR","created_at":"2026-06-10T01:11:08Z"},{"alias_kind":"pith_short_16","alias_value":"25LINL6LOXBRFLW5","created_at":"2026-06-10T01:11:08Z"},{"alias_kind":"pith_short_8","alias_value":"25LINL6L","created_at":"2026-06-10T01:11:08Z"}],"graph_snapshots":[{"event_id":"sha256:c11ff8ae8377023037f7373748e4e878c746c3dbe461864a1db237c5b32a1172","target":"graph","created_at":"2026-06-10T01:11:08Z","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/2606.11079/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Evaluation remains a critical bottleneck for interactive agent development. Existing evaluation methods often rely on static benchmarks, which fail to capture the dynamic, multi-step nature of agentic behavior and struggle to expose meaningful failure modes. While user-simulation-based evaluation offers a promising alternative, existing simulation frameworks suffer from two major limitations. First, they provide limited mechanisms for evaluating the quality and comprehensiveness of simulated interactions, making it difficult to assess whether a simulator sufficiently explores an agent's capabi","authors_text":"Ryan Shea, Yunan Lu, Yusen Zhang, Zhou Yu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-09T16:39:32Z","title":"VISTA: A Versatile Interactive User Simulation Toolkit for Agent Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.11079","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":"sha256:5bfce0976989dfd67981cd65c513e728f51a479494d1aa57c0bebbfc1ef2feac","target":"record","created_at":"2026-06-10T01:11:08Z","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":"3886f78d77a3d4b52228b02c1cbcdc4a75c50e7ba5434757f6896e9c5814de22","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-09T16:39:32Z","title_canon_sha256":"54cc92bbceba1bccbf0c6b5262dd0c63e374ab5bfc4d27989aebabb8717cd9c8"},"schema_version":"1.0","source":{"id":"2606.11079","kind":"arxiv","version":1}},"canonical_sha256":"d75686afcb75c312aedd4c6c8030151df0f5bc9d43587fd678451380198dd63c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d75686afcb75c312aedd4c6c8030151df0f5bc9d43587fd678451380198dd63c","first_computed_at":"2026-06-10T01:11:08.952716Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-10T01:11:08.952716Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fcQAcYETs9U3xAquwW6z5UNLKhgaSE4wFAtmKu39L5eTgpCLu1btn/1osXxdPy2h7WZENUEgknRhR9UgCGo5Ag==","signature_status":"signed_v1","signed_at":"2026-06-10T01:11:08.953625Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.11079","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5bfce0976989dfd67981cd65c513e728f51a479494d1aa57c0bebbfc1ef2feac","sha256:c11ff8ae8377023037f7373748e4e878c746c3dbe461864a1db237c5b32a1172"],"state_sha256":"d68905909ba519d8f1cd1bbdac11f29eb0f0d8140d25eb241a41136105de0117"}