{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:3XEF6LSLKJPUUSOG2ZDW5ZAKXM","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":"2197b5bdbbd4602d2935a6e78ff9af20b3c7de8f138286a4e8a71db6c6864cf4","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-09-13T16:31:44Z","title_canon_sha256":"ab661cb272cb43e5a9c859e8637cfd23926d5ccc6626b5adeef78828a91a6cf3"},"schema_version":"1.0","source":{"id":"1709.04409","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1709.04409","created_at":"2026-05-18T00:35:14Z"},{"alias_kind":"arxiv_version","alias_value":"1709.04409v1","created_at":"2026-05-18T00:35:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1709.04409","created_at":"2026-05-18T00:35:14Z"},{"alias_kind":"pith_short_12","alias_value":"3XEF6LSLKJPU","created_at":"2026-05-18T12:30:58Z"},{"alias_kind":"pith_short_16","alias_value":"3XEF6LSLKJPUUSOG","created_at":"2026-05-18T12:30:58Z"},{"alias_kind":"pith_short_8","alias_value":"3XEF6LSL","created_at":"2026-05-18T12:30:58Z"}],"graph_snapshots":[{"event_id":"sha256:d44c2958108760ca57733a44a84fad400a991505cc795bfca252f72120be80e7","target":"graph","created_at":"2026-05-18T00:35:14Z","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"},"paper":{"abstract_excerpt":"In contrast with goal-oriented dialogue, social dialogue has no clear measure of task success. Consequently, evaluation of these systems is notoriously hard. In this paper, we review current evaluation methods, focusing on automatic metrics. We conclude that turn-based metrics often ignore the context and do not account for the fact that several replies are valid, while end-of-dialogue rewards are mainly hand-crafted. Both lack grounding in human perceptions.","authors_text":"Amanda Cercas Curry, Helen Hastie, Verena Rieser","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-09-13T16:31:44Z","title":"A Review of Evaluation Techniques for Social Dialogue Systems"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1709.04409","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:d4852c7e2cad236e06317ef6f62b76917c40326a3410f14cdfa21877c7edef49","target":"record","created_at":"2026-05-18T00:35:14Z","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":"2197b5bdbbd4602d2935a6e78ff9af20b3c7de8f138286a4e8a71db6c6864cf4","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-09-13T16:31:44Z","title_canon_sha256":"ab661cb272cb43e5a9c859e8637cfd23926d5ccc6626b5adeef78828a91a6cf3"},"schema_version":"1.0","source":{"id":"1709.04409","kind":"arxiv","version":1}},"canonical_sha256":"ddc85f2e4b525f4a49c6d6476ee40abb0a005a526b48a2de46a86454b9e37d97","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ddc85f2e4b525f4a49c6d6476ee40abb0a005a526b48a2de46a86454b9e37d97","first_computed_at":"2026-05-18T00:35:14.445086Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:35:14.445086Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"o2GcukfJDf6LyQMApadpv7VmImMwVAt01GHnerl28r3Sy+ObnVvZThgDyJylyNLHX4WrWpsHSmcz0p32i0KdBQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:35:14.445751Z","signed_message":"canonical_sha256_bytes"},"source_id":"1709.04409","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d4852c7e2cad236e06317ef6f62b76917c40326a3410f14cdfa21877c7edef49","sha256:d44c2958108760ca57733a44a84fad400a991505cc795bfca252f72120be80e7"],"state_sha256":"1028fa91889a809b865df7cb4c14a9e57e153dfe437e7340d7c20f7d91ea0e34"}