{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:4I3W2RIPD5OCF6NDZU6ZO6FUFZ","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":"880697e2e3efe5aa520237cffbeba9ecc6a9307fac6146635dd111e1457855ea","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-09-30T12:01:29Z","title_canon_sha256":"fc9b01f24e57ab8c28b611cf9a4f4ec94e6f55a23ed7f9207d2acfdd43263abb"},"schema_version":"1.0","source":{"id":"2409.20222","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.20222","created_at":"2026-07-05T09:19:12Z"},{"alias_kind":"arxiv_version","alias_value":"2409.20222v2","created_at":"2026-07-05T09:19:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.20222","created_at":"2026-07-05T09:19:12Z"},{"alias_kind":"pith_short_12","alias_value":"4I3W2RIPD5OC","created_at":"2026-07-05T09:19:12Z"},{"alias_kind":"pith_short_16","alias_value":"4I3W2RIPD5OCF6ND","created_at":"2026-07-05T09:19:12Z"},{"alias_kind":"pith_short_8","alias_value":"4I3W2RIP","created_at":"2026-07-05T09:19:12Z"}],"graph_snapshots":[{"event_id":"sha256:87778f36f75be5cf76ad03b351fb2e3b33a50a84430f77cdd97dd3cf4d702693","target":"graph","created_at":"2026-07-05T09:19:12Z","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/2409.20222/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce a dynamic benchmarking system for conversational agents that evaluates their performance through a single, simulated, and lengthy user$\\leftrightarrow$agent interaction. The interaction is a conversation between the user and agent, where multiple tasks are introduced and then undertaken concurrently. We context switch regularly to interleave the tasks, which constructs a realistic testing scenario in which we assess the Long-Term Memory, Continual Learning, and Information Integration capabilities of the agents. Results from both proprietary and open-source Large-Language Models s","authors_text":"David Castillo-Bolado, Finlay Gray, Joseph Davidson, Marek Rosa","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-09-30T12:01:29Z","title":"Beyond Prompts: Dynamic Conversational Benchmarking of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.20222","kind":"arxiv","version":2},"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":"sha256:57f2a9fe873aa8fb265d5dca9de6a89a53e5ceb868654b15212b8978f1038b62","target":"record","created_at":"2026-07-05T09:19:12Z","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":"880697e2e3efe5aa520237cffbeba9ecc6a9307fac6146635dd111e1457855ea","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-09-30T12:01:29Z","title_canon_sha256":"fc9b01f24e57ab8c28b611cf9a4f4ec94e6f55a23ed7f9207d2acfdd43263abb"},"schema_version":"1.0","source":{"id":"2409.20222","kind":"arxiv","version":2}},"canonical_sha256":"e2376d450f1f5c22f9a3cd3d9778b42e721238e216b1b771094a51ad6ea4714f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e2376d450f1f5c22f9a3cd3d9778b42e721238e216b1b771094a51ad6ea4714f","first_computed_at":"2026-07-05T09:19:12.537141Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:19:12.537141Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vRMhpl1+vEuYEOzW3e82Bear+PwbMdO/y0+OM4U2GRAGza3ikCKIpwHL3WWJWCTHaSVV1NhPL5kWddcVBOWLCA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:19:12.537641Z","signed_message":"canonical_sha256_bytes"},"source_id":"2409.20222","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:57f2a9fe873aa8fb265d5dca9de6a89a53e5ceb868654b15212b8978f1038b62","sha256:87778f36f75be5cf76ad03b351fb2e3b33a50a84430f77cdd97dd3cf4d702693"],"state_sha256":"43a45dc14697d50dad0a18a000589e45b00dfbd2890a79fe64e6fad03f6bd6e2"}