{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:BOUSXS2SP4QUXN3JFQKUC4MQ4M","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":"336b196c1c98f8b3f27e51c7407461a54ccc4626502d1aba01a7f4fe9f8d99ba","cross_cats_sorted":["stat.AP"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-22T16:08:44Z","title_canon_sha256":"437a9e97c6513aab96ed5e2db83eb587dc595f0f556f6826863d742824911608"},"schema_version":"1.0","source":{"id":"2506.18082","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.18082","created_at":"2026-07-05T11:26:06Z"},{"alias_kind":"arxiv_version","alias_value":"2506.18082v2","created_at":"2026-07-05T11:26:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.18082","created_at":"2026-07-05T11:26:06Z"},{"alias_kind":"pith_short_12","alias_value":"BOUSXS2SP4QU","created_at":"2026-07-05T11:26:06Z"},{"alias_kind":"pith_short_16","alias_value":"BOUSXS2SP4QUXN3J","created_at":"2026-07-05T11:26:06Z"},{"alias_kind":"pith_short_8","alias_value":"BOUSXS2S","created_at":"2026-07-05T11:26:06Z"}],"graph_snapshots":[{"event_id":"sha256:6c9361b7a8ab27e0ad469c4cfdd7827af2c952be9c44c7964583f24a9e6258fe","target":"graph","created_at":"2026-07-05T11:26:06Z","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/2506.18082/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Assessing the quality of LLM-generated text remains a fundamental challenge in natural language processing. Current evaluation approaches often rely on isolated metrics or simplistic aggregations that fail to capture the nuanced trade-offs between coherence, diversity, fluency, and other relevant indicators of text quality. In this work, we adapt a recently proposed framework for statistical inference based on Generalized Stochastic Dominance (GSD) that addresses three critical limitations in existing benchmarking methodologies: the inadequacy of single-metric evaluation, the incompatibility b","authors_text":"Christoph Jansen, Esteban Garces Arias, Hannah Blocher, Julian Rodemann, Matthias A{\\ss}enmacher","cross_cats":["stat.AP"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-22T16:08:44Z","title":"Statistical Multicriteria Evaluation of LLM-Generated Text"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.18082","kind":"arxiv","version":2},"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":"sha256:45dd888e302fb78f4c9e9cc1b185acfbd24b13acf609e87d8612af82f2e7c0d9","target":"record","created_at":"2026-07-05T11:26:06Z","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":"336b196c1c98f8b3f27e51c7407461a54ccc4626502d1aba01a7f4fe9f8d99ba","cross_cats_sorted":["stat.AP"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-22T16:08:44Z","title_canon_sha256":"437a9e97c6513aab96ed5e2db83eb587dc595f0f556f6826863d742824911608"},"schema_version":"1.0","source":{"id":"2506.18082","kind":"arxiv","version":2}},"canonical_sha256":"0ba92bcb527f214bb7692c15417190e30aa97135bd09431ab471cf77e2887cfe","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0ba92bcb527f214bb7692c15417190e30aa97135bd09431ab471cf77e2887cfe","first_computed_at":"2026-07-05T11:26:06.130157Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:26:06.130157Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uKgudZV+OhfWxxDReAJusmL/1OIzZ98lI9lB5V9S0TsL2AjF2InLaPCJKfLNm9AnNbULfMO8p2pSU2KwW6kNCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:26:06.130644Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.18082","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:45dd888e302fb78f4c9e9cc1b185acfbd24b13acf609e87d8612af82f2e7c0d9","sha256:6c9361b7a8ab27e0ad469c4cfdd7827af2c952be9c44c7964583f24a9e6258fe"],"state_sha256":"baaed924fc54b4c80b90c3389f5f05fc2502db61db842163cffddb5903401cd5"}