{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:EWU56QZOKNWZFMEPYF57EBP24I","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":"f879c682e88cfcca9787bc928f11e784f26ed2d3ab35aff05ce51ffa630d177e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-29T22:16:39Z","title_canon_sha256":"6c276c5dcd3efc11f2b312fac21543011ac9f126b9acb49de688cd098351919d"},"schema_version":"1.0","source":{"id":"2504.21202","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.21202","created_at":"2026-07-05T10:56:19Z"},{"alias_kind":"arxiv_version","alias_value":"2504.21202v1","created_at":"2026-07-05T10:56:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.21202","created_at":"2026-07-05T10:56:19Z"},{"alias_kind":"pith_short_12","alias_value":"EWU56QZOKNWZ","created_at":"2026-07-05T10:56:19Z"},{"alias_kind":"pith_short_16","alias_value":"EWU56QZOKNWZFMEP","created_at":"2026-07-05T10:56:19Z"},{"alias_kind":"pith_short_8","alias_value":"EWU56QZO","created_at":"2026-07-05T10:56:19Z"}],"graph_snapshots":[{"event_id":"sha256:5caefcd4fcd1c44f99e411929efebc231bcee2aeb2c51bf212a92a5170d6f23d","target":"graph","created_at":"2026-07-05T10:56:19Z","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/2504.21202/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Despite the recent advances in Large Language Models, benchmarks for evaluating legal writing remain scarce due to the inherent complexity of assessing open-ended responses in this domain. One of the key challenges in evaluating language models on domain-specific tasks is finding test datasets that are public, frequently updated, and contain comprehensive evaluation guidelines. The Brazilian Bar Examination meets these requirements. We introduce oab-bench, a benchmark comprising 105 questions across seven areas of law from recent editions of the exam. The benchmark includes comprehensive evalu","authors_text":"Ramon Pires, Rodrigo Nogueira, Roseval Malaquias Junior","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-29T22:16:39Z","title":"Automatic Legal Writing Evaluation of LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.21202","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:cfed5867e352d31347b19e94178e379d0c9794e49b9e9f7dbba774c3934b2296","target":"record","created_at":"2026-07-05T10:56:19Z","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":"f879c682e88cfcca9787bc928f11e784f26ed2d3ab35aff05ce51ffa630d177e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-29T22:16:39Z","title_canon_sha256":"6c276c5dcd3efc11f2b312fac21543011ac9f126b9acb49de688cd098351919d"},"schema_version":"1.0","source":{"id":"2504.21202","kind":"arxiv","version":1}},"canonical_sha256":"25a9df432e536d92b08fc17bf205fae20bca2d902b8d6f297b38afe150fdb4d1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"25a9df432e536d92b08fc17bf205fae20bca2d902b8d6f297b38afe150fdb4d1","first_computed_at":"2026-07-05T10:56:19.244393Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:56:19.244393Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"jkJoUuvEuJXhfN+W7ddSwHBreQZCvyOlrMX6u7s5fhL3UFodKNaeqD1oD0fECmQuFijKshyuXUWpIs2JtrgrAw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:56:19.244930Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.21202","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cfed5867e352d31347b19e94178e379d0c9794e49b9e9f7dbba774c3934b2296","sha256:5caefcd4fcd1c44f99e411929efebc231bcee2aeb2c51bf212a92a5170d6f23d"],"state_sha256":"70e82532082d65fb0f1ced045e3dd020e075c31250350dc900c6e7993c6db9e7"}