{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:XT5TA4L47A3MTN2CHAKW6LVD6B","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":"d78c3f1185ebe61fa1ee1c00087b1ebdc12e4b8bff6750b30ada8b7b6c13b8fd","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-12T07:44:36Z","title_canon_sha256":"2882b4a900f1c414db1c33f7825806179f85af6500e8c8ee56b32bd239f87169"},"schema_version":"1.0","source":{"id":"2406.07967","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.07967","created_at":"2026-07-05T08:30:41Z"},{"alias_kind":"arxiv_version","alias_value":"2406.07967v1","created_at":"2026-07-05T08:30:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.07967","created_at":"2026-07-05T08:30:41Z"},{"alias_kind":"pith_short_12","alias_value":"XT5TA4L47A3M","created_at":"2026-07-05T08:30:41Z"},{"alias_kind":"pith_short_16","alias_value":"XT5TA4L47A3MTN2C","created_at":"2026-07-05T08:30:41Z"},{"alias_kind":"pith_short_8","alias_value":"XT5TA4L4","created_at":"2026-07-05T08:30:41Z"}],"graph_snapshots":[{"event_id":"sha256:bbb4126418abc20b90e7f00bff105c81c672fed6c368879857a593b3c65aca52","target":"graph","created_at":"2026-07-05T08:30:41Z","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/2406.07967/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Human evaluation is viewed as a reliable evaluation method for NLG which is expensive and time-consuming. To save labor and costs, researchers usually perform human evaluation on a small subset of data sampled from the whole dataset in practice. However, different selection subsets will lead to different rankings of the systems. To give a more correct inter-system ranking and make the gold standard human evaluation more reliable, we propose a Constrained Active Sampling Framework (CASF) for reliable human judgment. CASF operates through a Learner, a Systematic Sampler and a Constrained Control","authors_text":"Jie Ruan, Mingqi Gao, Xiaojun Wan, Xiao Pu, Yuesheng Zhu","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-12T07:44:36Z","title":"Better than Random: Reliable NLG Human Evaluation with Constrained Active Sampling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.07967","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":"sha256:57ddc7e08891670cc566b6a57d767c3f9e9d46a3e620e285fcc455494b6854a0","target":"record","created_at":"2026-07-05T08:30:41Z","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":"d78c3f1185ebe61fa1ee1c00087b1ebdc12e4b8bff6750b30ada8b7b6c13b8fd","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-12T07:44:36Z","title_canon_sha256":"2882b4a900f1c414db1c33f7825806179f85af6500e8c8ee56b32bd239f87169"},"schema_version":"1.0","source":{"id":"2406.07967","kind":"arxiv","version":1}},"canonical_sha256":"bcfb30717cf836c9b74238156f2ea3f07489722e0fe7bdfaa49c600d0a763524","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bcfb30717cf836c9b74238156f2ea3f07489722e0fe7bdfaa49c600d0a763524","first_computed_at":"2026-07-05T08:30:41.413361Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:30:41.413361Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zsgsnwBN/jnWdkEtjQWh6ykN3ljYWh3C7JeyzRPpU45oZcNPYeUsXy701XXBS4XA4SCjff0GFRNO1TvQODIFCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:30:41.413854Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.07967","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:57ddc7e08891670cc566b6a57d767c3f9e9d46a3e620e285fcc455494b6854a0","sha256:bbb4126418abc20b90e7f00bff105c81c672fed6c368879857a593b3c65aca52"],"state_sha256":"485b3a294485b2452186b34305ff8c5d0289c4fe31bbb7b39af04a1aef7282d3"}