{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:GNCCMZXRBFODL4PHJKTZA6Y732","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":"f90d6f0e1897c3c093954d6c3003607af2778d507d592faaa290af2d9221a885","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-09T07:27:15Z","title_canon_sha256":"cae87f51c5a87cab213c557b763b7f03cadbd27c928532318d17eb167afdb073"},"schema_version":"1.0","source":{"id":"2310.05470","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.05470","created_at":"2026-07-05T07:21:28Z"},{"alias_kind":"arxiv_version","alias_value":"2310.05470v2","created_at":"2026-07-05T07:21:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.05470","created_at":"2026-07-05T07:21:28Z"},{"alias_kind":"pith_short_12","alias_value":"GNCCMZXRBFOD","created_at":"2026-07-05T07:21:28Z"},{"alias_kind":"pith_short_16","alias_value":"GNCCMZXRBFODL4PH","created_at":"2026-07-05T07:21:28Z"},{"alias_kind":"pith_short_8","alias_value":"GNCCMZXR","created_at":"2026-07-05T07:21:28Z"}],"graph_snapshots":[{"event_id":"sha256:9beb400312ad2538a18b7611b0dbbc5b5218c3fb29914e6f72b43763952755ea","target":"graph","created_at":"2026-07-05T07:21:28Z","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/2310.05470/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The rapid development of Large Language Models (LLMs) has substantially expanded the range of tasks they can address. In the field of Natural Language Processing (NLP), researchers have shifted their focus from conventional NLP tasks (e.g., sequence tagging and parsing) towards tasks that revolve around aligning with human needs (e.g., brainstorming and email writing). This shift in task distribution imposes new requirements on evaluating these aligned models regarding generality (i.e., assessing performance across diverse scenarios), flexibility (i.e., examining under different protocols), an","authors_text":"Hai Zhao, Junlong Li, Pengfei Liu, Run-Ze Fan, Shichao Sun, Weizhe Yuan","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-09T07:27:15Z","title":"Generative Judge for Evaluating Alignment"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.05470","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:faa51fe81c3fc98049fe15371b58a12d599aade10cbdcb0f0395298f67b12d75","target":"record","created_at":"2026-07-05T07:21:28Z","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":"f90d6f0e1897c3c093954d6c3003607af2778d507d592faaa290af2d9221a885","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-09T07:27:15Z","title_canon_sha256":"cae87f51c5a87cab213c557b763b7f03cadbd27c928532318d17eb167afdb073"},"schema_version":"1.0","source":{"id":"2310.05470","kind":"arxiv","version":2}},"canonical_sha256":"33442666f1095c35f1e74aa7907b1fde817a2382c495d04b22ca6f45f26407a1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"33442666f1095c35f1e74aa7907b1fde817a2382c495d04b22ca6f45f26407a1","first_computed_at":"2026-07-05T07:21:28.388682Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:21:28.388682Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6lqVvyhrRL+gMrZdwgWpRdEuJi7j7cZG2WNd87PaldwM1HbBhWbpMY4yamjLz5rs2poLW9M44BZoyHFfsbWcDg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:21:28.389165Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.05470","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:faa51fe81c3fc98049fe15371b58a12d599aade10cbdcb0f0395298f67b12d75","sha256:9beb400312ad2538a18b7611b0dbbc5b5218c3fb29914e6f72b43763952755ea"],"state_sha256":"e4edc5ab8dd31e5e7d8fce452ba46660ae65bbc3e49ee2eee0d16a5721abaf4d"}