{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:N6GK2YSG4RF4LRUYN44VZK5DNF","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":"4893df0db0315c18889831ab504d1138d8cb150e7e6a504c14fa7deb6800113f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-06-10T23:29:05Z","title_canon_sha256":"bdacdeea2d27f0606c88e12eed58796b5fb8451964859804f0dc96bc53e6819d"},"schema_version":"1.0","source":{"id":"2006.06110","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2006.06110","created_at":"2026-07-05T01:09:21Z"},{"alias_kind":"arxiv_version","alias_value":"2006.06110v1","created_at":"2026-07-05T01:09:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.06110","created_at":"2026-07-05T01:09:21Z"},{"alias_kind":"pith_short_12","alias_value":"N6GK2YSG4RF4","created_at":"2026-07-05T01:09:21Z"},{"alias_kind":"pith_short_16","alias_value":"N6GK2YSG4RF4LRUY","created_at":"2026-07-05T01:09:21Z"},{"alias_kind":"pith_short_8","alias_value":"N6GK2YSG","created_at":"2026-07-05T01:09:21Z"}],"graph_snapshots":[{"event_id":"sha256:2124693a262899f76b3a34e747c6d2ee8b813024b68a192e753838c57e832bd0","target":"graph","created_at":"2026-07-05T01:09:21Z","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/2006.06110/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As conversational AI-based dialogue management has increasingly become a trending topic, the need for a standardized and reliable evaluation procedure grows even more pressing. The current state of affairs suggests various evaluation protocols to assess chat-oriented dialogue management systems, rendering it difficult to conduct fair comparative studies across different approaches and gain an insightful understanding of their values. To foster this research, a more robust evaluation protocol must be set in place. This paper presents a comprehensive synthesis of both automated and human evaluat","authors_text":"Jinho D. Choi, Sarah E. Finch","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-06-10T23:29:05Z","title":"Towards Unified Dialogue System Evaluation: A Comprehensive Analysis of Current Evaluation Protocols"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.06110","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":"sha256:9627e603e218d2a9eeac0122f64ea2a6c113ab8c608c3a778e576d01574ba973","target":"record","created_at":"2026-07-05T01:09:21Z","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":"4893df0db0315c18889831ab504d1138d8cb150e7e6a504c14fa7deb6800113f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-06-10T23:29:05Z","title_canon_sha256":"bdacdeea2d27f0606c88e12eed58796b5fb8451964859804f0dc96bc53e6819d"},"schema_version":"1.0","source":{"id":"2006.06110","kind":"arxiv","version":1}},"canonical_sha256":"6f8cad6246e44bc5c6986f395caba3696834c57a698ec6dc69ce0432629986e6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6f8cad6246e44bc5c6986f395caba3696834c57a698ec6dc69ce0432629986e6","first_computed_at":"2026-07-05T01:09:21.256470Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:09:21.256470Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fADn3n2LjI3p5E88EMTfxsppAZ2y0OvawEpTJ4BnZ7d/PdXPlcWpzNVcPpJH7Qrodcs5J5x+JZg1PdtWtCgXDw==","signature_status":"signed_v1","signed_at":"2026-07-05T01:09:21.256905Z","signed_message":"canonical_sha256_bytes"},"source_id":"2006.06110","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9627e603e218d2a9eeac0122f64ea2a6c113ab8c608c3a778e576d01574ba973","sha256:2124693a262899f76b3a34e747c6d2ee8b813024b68a192e753838c57e832bd0"],"state_sha256":"4745fe4e791e845ef53eaf80bb084ebed2c50e58dda8b3914579886b30b445b4"}