{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:EACVKCNYYVHWGARNL5JQHLZOUN","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":"555c5a3f933fa0e268011c817409d47a2f2b47637a4c338ac3dd965d8b16e4bb","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T05:57:09Z","title_canon_sha256":"87bb23a1c2d723e05960498bfc7494099654e91dcaf01790d335e4108fc8065e"},"schema_version":"1.0","source":{"id":"2305.13711","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.13711","created_at":"2026-07-05T06:12:56Z"},{"alias_kind":"arxiv_version","alias_value":"2305.13711v1","created_at":"2026-07-05T06:12:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.13711","created_at":"2026-07-05T06:12:56Z"},{"alias_kind":"pith_short_12","alias_value":"EACVKCNYYVHW","created_at":"2026-07-05T06:12:56Z"},{"alias_kind":"pith_short_16","alias_value":"EACVKCNYYVHWGARN","created_at":"2026-07-05T06:12:56Z"},{"alias_kind":"pith_short_8","alias_value":"EACVKCNY","created_at":"2026-07-05T06:12:56Z"}],"graph_snapshots":[{"event_id":"sha256:50dcf1397d66ea9a35f4e6fdfb66ab4595214636c38d9de660bcfcad74b37409","target":"graph","created_at":"2026-07-05T06:12:56Z","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/2305.13711/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We propose LLM-Eval, a unified multi-dimensional automatic evaluation method for open-domain conversations with large language models (LLMs). Existing evaluation methods often rely on human annotations, ground-truth responses, or multiple LLM prompts, which can be expensive and time-consuming. To address these issues, we design a single prompt-based evaluation method that leverages a unified evaluation schema to cover multiple dimensions of conversation quality in a single model call. We extensively evaluate the performance of LLM-Eval on various benchmark datasets, demonstrating its effective","authors_text":"Yen-Ting Lin, Yun-Nung Chen","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T05:57:09Z","title":"LLM-Eval: Unified Multi-Dimensional Automatic Evaluation for Open-Domain Conversations with Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.13711","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:a5c82dedf5063ed41aa176b02d325cd72c6b9e3f5dbd20203f39cad754e03639","target":"record","created_at":"2026-07-05T06:12:56Z","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":"555c5a3f933fa0e268011c817409d47a2f2b47637a4c338ac3dd965d8b16e4bb","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T05:57:09Z","title_canon_sha256":"87bb23a1c2d723e05960498bfc7494099654e91dcaf01790d335e4108fc8065e"},"schema_version":"1.0","source":{"id":"2305.13711","kind":"arxiv","version":1}},"canonical_sha256":"20055509b8c54f63022d5f5303af2ea3770b548d3237d55a79f2d7d4e115ff97","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"20055509b8c54f63022d5f5303af2ea3770b548d3237d55a79f2d7d4e115ff97","first_computed_at":"2026-07-05T06:12:56.518253Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:12:56.518253Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Vw1Xsgd9mSTezp5fQW7s6LWmIe+lHyc8NOhPg1q+uQbQRiNmVnZDvr9v36vSiJlGERU3irSUNu/rFokbYiveBw==","signature_status":"signed_v1","signed_at":"2026-07-05T06:12:56.518679Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.13711","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a5c82dedf5063ed41aa176b02d325cd72c6b9e3f5dbd20203f39cad754e03639","sha256:50dcf1397d66ea9a35f4e6fdfb66ab4595214636c38d9de660bcfcad74b37409"],"state_sha256":"7305a36b4fd881fd61b4da0649177e8321932c3e02e91cb86dbf2a3466daea40"}