{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:HSVQEW3WEYNFABRHQ76MBAPTDD","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":"fd43e6a95dc76bbadd7656cd693b987c0899c506e1b5dd01ad333ffb805fd4ca","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-12-19T18:59:45Z","title_canon_sha256":"288d610b49d327f58d06a470a35eaa5e9b2d0dc0d1c0229b0897ec3abab066d5"},"schema_version":"1.0","source":{"id":"2212.09746","kind":"arxiv","version":5}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2212.09746","created_at":"2026-07-05T07:30:45Z"},{"alias_kind":"arxiv_version","alias_value":"2212.09746v5","created_at":"2026-07-05T07:30:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.09746","created_at":"2026-07-05T07:30:45Z"},{"alias_kind":"pith_short_12","alias_value":"HSVQEW3WEYNF","created_at":"2026-07-05T07:30:45Z"},{"alias_kind":"pith_short_16","alias_value":"HSVQEW3WEYNFABRH","created_at":"2026-07-05T07:30:45Z"},{"alias_kind":"pith_short_8","alias_value":"HSVQEW3W","created_at":"2026-07-05T07:30:45Z"}],"graph_snapshots":[{"event_id":"sha256:ddc51e7a1f3c48f93308496b54d5078fa770e7aa9261b828dd64d699458e972b","target":"graph","created_at":"2026-07-05T07:30:45Z","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/2212.09746/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Many real-world applications of language models (LMs), such as writing assistance and code autocomplete, involve human-LM interaction. However, most benchmarks are non-interactive in that a model produces output without human involvement. To evaluate human-LM interaction, we develop a new framework, Human-AI Language-based Interaction Evaluation (HALIE), that defines the components of interactive systems and dimensions to consider when designing evaluation metrics. Compared to standard, non-interactive evaluation, HALIE captures (i) the interactive process, not only the final output; (ii) the ","authors_text":"Amelia Hardy, Ashwin Paranjape, Esin Durmus, Faisal Ladhak, Frieda Rong, Hancheng Cao, Ines Gerard-Ursin, John Thickstun, Joon Sung Park, Megha Srivastava, Michael Bernstein, Minae Kwon, Mina Lee, Percy Liang, Rishi Bommasani, Rose E. Wang, Tony Lee, Xiang Lisa Li","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-12-19T18:59:45Z","title":"Evaluating Human-Language Model Interaction"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.09746","kind":"arxiv","version":5},"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":"sha256:15f66739b01a6374f9e8b2f36a61ec5f1c59f9f395753cf6be8470d554e412f0","target":"record","created_at":"2026-07-05T07:30:45Z","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":"fd43e6a95dc76bbadd7656cd693b987c0899c506e1b5dd01ad333ffb805fd4ca","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-12-19T18:59:45Z","title_canon_sha256":"288d610b49d327f58d06a470a35eaa5e9b2d0dc0d1c0229b0897ec3abab066d5"},"schema_version":"1.0","source":{"id":"2212.09746","kind":"arxiv","version":5}},"canonical_sha256":"3cab025b76261a50062787fcc081f318cc398cd2a7c823366a636c886abbb834","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3cab025b76261a50062787fcc081f318cc398cd2a7c823366a636c886abbb834","first_computed_at":"2026-07-05T07:30:45.974702Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:30:45.974702Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"kDHUWH8MXEFK3HXSHUKowNQyNP345Cm07L4g3z6gnxdD/zCq9tNBUV8XTUI+vJ2i8sHU9zWsR8LVqfOTiKyKAA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:30:45.975159Z","signed_message":"canonical_sha256_bytes"},"source_id":"2212.09746","source_kind":"arxiv","source_version":5}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:15f66739b01a6374f9e8b2f36a61ec5f1c59f9f395753cf6be8470d554e412f0","sha256:ddc51e7a1f3c48f93308496b54d5078fa770e7aa9261b828dd64d699458e972b"],"state_sha256":"c46e3dec02dd31ada0d6ac8beac052d702d27d76676dd8aeffafeff5fb42bd6d"}