{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:LDSFXRLG2UMWXKBMZISHFMU6SA","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":"9feff07b3272742b5c8c979a680483adc5b86f2688f31bc57765d9a3af673d1b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-24T20:32:49Z","title_canon_sha256":"90f77d6bf8e18b1e105cd636bc37e3caea40657cea173f5fbf3e5c81916cccf2"},"schema_version":"1.0","source":{"id":"2405.15924","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.15924","created_at":"2026-07-05T08:25:04Z"},{"alias_kind":"arxiv_version","alias_value":"2405.15924v3","created_at":"2026-07-05T08:25:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.15924","created_at":"2026-07-05T08:25:04Z"},{"alias_kind":"pith_short_12","alias_value":"LDSFXRLG2UMW","created_at":"2026-07-05T08:25:04Z"},{"alias_kind":"pith_short_16","alias_value":"LDSFXRLG2UMWXKBM","created_at":"2026-07-05T08:25:04Z"},{"alias_kind":"pith_short_8","alias_value":"LDSFXRLG","created_at":"2026-07-05T08:25:04Z"}],"graph_snapshots":[{"event_id":"sha256:bc098d9e6832ba944bdd3ffaa3776ee4ce30c10b890e7b019952718ca6cd27c2","target":"graph","created_at":"2026-07-05T08:25:04Z","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/2405.15924/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The long-standing one-to-many problem of gold standard responses in open-domain dialogue systems presents challenges for automatic evaluation metrics. Though prior works have demonstrated some success by applying powerful Large Language Models (LLMs), existing approaches still struggle with the one-to-many problem, and exhibit subpar performance in domain-specific scenarios. We assume the commonsense reasoning biases within LLMs may hinder their performance in domainspecific evaluations. To address both issues, we propose a novel framework SLIDE (Small and Large Integrated for Dialogue Evaluat","authors_text":"Bohao Yang, Chenghua Lin, Chen Tang, Kun Zhao, Liang Zhan","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-24T20:32:49Z","title":"SLIDE: A Framework Integrating Small and Large Language Models for Open-Domain Dialogues Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.15924","kind":"arxiv","version":3},"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":"sha256:3bf914488dad08e22f7b718d0f64e67f0ea052c12ec9e23a06b56882005183d8","target":"record","created_at":"2026-07-05T08:25:04Z","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":"9feff07b3272742b5c8c979a680483adc5b86f2688f31bc57765d9a3af673d1b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-24T20:32:49Z","title_canon_sha256":"90f77d6bf8e18b1e105cd636bc37e3caea40657cea173f5fbf3e5c81916cccf2"},"schema_version":"1.0","source":{"id":"2405.15924","kind":"arxiv","version":3}},"canonical_sha256":"58e45bc566d5196ba82cca2472b29e901c8c63a7d8a1f5b6b435e5ed0b3f8bbf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"58e45bc566d5196ba82cca2472b29e901c8c63a7d8a1f5b6b435e5ed0b3f8bbf","first_computed_at":"2026-07-05T08:25:04.893420Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:25:04.893420Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"OWqmG6SeD75LK/lkqOZzeKlcQuSSigaGr8wmGgfa9dlU3eDowz4005bZJW4QKxycbU0fKb9iAsbgcmVv8QXECg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:25:04.893951Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.15924","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3bf914488dad08e22f7b718d0f64e67f0ea052c12ec9e23a06b56882005183d8","sha256:bc098d9e6832ba944bdd3ffaa3776ee4ce30c10b890e7b019952718ca6cd27c2"],"state_sha256":"0b7b855f5be4de75aa7ac34ac3d5d32a911e2c3305792c15f23ae2a608d6e3cb"}