{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EWU56QZOKNWZFMEPYF57EBP24I","short_pith_number":"pith:EWU56QZO","schema_version":"1.0","canonical_sha256":"25a9df432e536d92b08fc17bf205fae20bca2d902b8d6f297b38afe150fdb4d1","source":{"kind":"arxiv","id":"2504.21202","version":1},"attestation_state":"computed","paper":{"title":"Automatic Legal Writing Evaluation of LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ramon Pires, Rodrigo Nogueira, Roseval Malaquias Junior","submitted_at":"2025-04-29T22:16:39Z","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"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2504.21202","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-29T22:16:39Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6c276c5dcd3efc11f2b312fac21543011ac9f126b9acb49de688cd098351919d","abstract_canon_sha256":"f879c682e88cfcca9787bc928f11e784f26ed2d3ab35aff05ce51ffa630d177e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:56:19.244930Z","signature_b64":"jkJoUuvEuJXhfN+W7ddSwHBreQZCvyOlrMX6u7s5fhL3UFodKNaeqD1oD0fECmQuFijKshyuXUWpIs2JtrgrAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"25a9df432e536d92b08fc17bf205fae20bca2d902b8d6f297b38afe150fdb4d1","last_reissued_at":"2026-07-05T10:56:19.244393Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:56:19.244393Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Automatic Legal Writing Evaluation of LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ramon Pires, Rodrigo Nogueira, Roseval Malaquias Junior","submitted_at":"2025-04-29T22:16:39Z","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"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.21202","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2504.21202/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2504.21202","created_at":"2026-07-05T10:56:19.244451+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.21202v1","created_at":"2026-07-05T10:56:19.244451+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.21202","created_at":"2026-07-05T10:56:19.244451+00:00"},{"alias_kind":"pith_short_12","alias_value":"EWU56QZOKNWZ","created_at":"2026-07-05T10:56:19.244451+00:00"},{"alias_kind":"pith_short_16","alias_value":"EWU56QZOKNWZFMEP","created_at":"2026-07-05T10:56:19.244451+00:00"},{"alias_kind":"pith_short_8","alias_value":"EWU56QZO","created_at":"2026-07-05T10:56:19.244451+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22723","citing_title":"BLUEX v2: Benchmarking LLMs on Open-Ended Questions from Brazilian University Entrance Exams","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.22723","citing_title":"BLUEX v2: Benchmarking LLMs on Open-Ended Questions from Brazilian University Entrance Exams","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01630","citing_title":"Prosa: Rubric-Based Evaluation of LLMs on Real User Chats in Brazilian Portuguese","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EWU56QZOKNWZFMEPYF57EBP24I","json":"https://pith.science/pith/EWU56QZOKNWZFMEPYF57EBP24I.json","graph_json":"https://pith.science/api/pith-number/EWU56QZOKNWZFMEPYF57EBP24I/graph.json","events_json":"https://pith.science/api/pith-number/EWU56QZOKNWZFMEPYF57EBP24I/events.json","paper":"https://pith.science/paper/EWU56QZO"},"agent_actions":{"view_html":"https://pith.science/pith/EWU56QZOKNWZFMEPYF57EBP24I","download_json":"https://pith.science/pith/EWU56QZOKNWZFMEPYF57EBP24I.json","view_paper":"https://pith.science/paper/EWU56QZO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.21202&json=true","fetch_graph":"https://pith.science/api/pith-number/EWU56QZOKNWZFMEPYF57EBP24I/graph.json","fetch_events":"https://pith.science/api/pith-number/EWU56QZOKNWZFMEPYF57EBP24I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EWU56QZOKNWZFMEPYF57EBP24I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EWU56QZOKNWZFMEPYF57EBP24I/action/storage_attestation","attest_author":"https://pith.science/pith/EWU56QZOKNWZFMEPYF57EBP24I/action/author_attestation","sign_citation":"https://pith.science/pith/EWU56QZOKNWZFMEPYF57EBP24I/action/citation_signature","submit_replication":"https://pith.science/pith/EWU56QZOKNWZFMEPYF57EBP24I/action/replication_record"}},"created_at":"2026-07-05T10:56:19.244451+00:00","updated_at":"2026-07-05T10:56:19.244451+00:00"}