{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,20]],"date-time":"2026-06-20T16:49:44Z","timestamp":1781974184267,"version":"3.54.5"},"reference-count":55,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2025,1,1]],"date-time":"2025-01-01T00:00:00Z","timestamp":1735689600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"name":"Google Tensor Processing Unit (TPU) Research Cloud Program"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Access"],"published-print":{"date-parts":[[2025]]},"DOI":"10.1109\/access.2025.3645771","type":"journal-article","created":{"date-parts":[[2025,12,18]],"date-time":"2025-12-18T18:34:32Z","timestamp":1766082872000},"page":"214180-214200","source":"Crossref","is-referenced-by-count":1,"title":["PoETa v2: Toward More Robust Evaluation of Large Language Models in Portuguese"],"prefix":"10.1109","volume":"13","author":[{"given":"Thales","family":"Rog\u00e9rio Sales Almeida","sequence":"first","affiliation":[{"name":"Institute of Computing, University of Campinas, Campinas, São Paulo, Brazil"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0023-1971","authenticated-orcid":false,"given":"Ramon","family":"Pires","sequence":"additional","affiliation":[{"name":"Institute of Computing, University of Campinas, Campinas, São Paulo, Brazil"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hugo","family":"Abonizio","sequence":"additional","affiliation":[{"name":"Maritaca AI, Campinas, São Paulo, Brazil"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rodrigo","family":"Nogueira","sequence":"additional","affiliation":[{"name":"Maritaca AI, Campinas, São Paulo, Brazil"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0125-630X","authenticated-orcid":false,"given":"Helio","family":"Pedrini","sequence":"additional","affiliation":[{"name":"Institute of Computing, University of Campinas, Campinas, São Paulo, Brazil"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/HONET63146.2024.10822885"},{"key":"ref2","article-title":"A complete survey on LLM-based AI chatbots","author":"Dam","year":"2024","journal-title":"arXiv:2406.16937"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2025.3533217"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-025-02954-4"},{"key":"ref5","article-title":"TiEBe: Tracking language model recall of notable worldwide events through time","author":"Almeida","year":"2025","journal-title":"arXiv:2501.07482"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1145\/3630106.3658967"},{"key":"ref7","first-page":"78104","article-title":"BLEnD: A benchmark for LLMs on everyday knowledge in diverse cultures and languages","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"37","author":"Myung"},{"key":"ref8","article-title":"Bridging the data provenance gap across text, speech and video","author":"Longpre","year":"2024","journal-title":"arXiv:2412.17847"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-45392-2_15"},{"key":"ref10","article-title":"Sabi\u00e1-2: A new generation of Portuguese large language models","author":"Almeida","year":"2024","journal-title":"arXiv:2403.09887"},{"key":"ref11","article-title":"Sabi\u00e1-3 technical report","volume-title":"arXiv:2410.12049","author":"Abonizio","year":"2024"},{"key":"ref12","article-title":"Cabrita: Closing the gap for foreign languages","author":"Larcher","year":"2023","journal-title":"arXiv:2308.11878"},{"key":"ref13","article-title":"Gl\u00f3rIA\u2014A generative and open large language model for Portuguese","author":"Lopes","year":"2024","journal-title":"arXiv:2402. 12969"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1016\/j.mlwa.2024.100558"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1016\/j.patter.2025.101325"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.5753\/jbcs.2025.5788"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/BRACIS.2018.00016"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-45368-7_22"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.3233\/978-1-61499-838-9-89"},{"key":"ref20","article-title":"Automatic legal writing evaluation of LLMs","author":"Pires","year":"2025","journal-title":"arXiv:2504.21202"},{"key":"ref21","article-title":"Building a sentiment corpus of tweets in Brazilian Portuguese","author":"Brum","year":"2017","journal-title":"arXiv:1712.08917"},{"key":"ref22","article-title":"HateBR: A large expert annotated corpus of Brazilian Instagram comments for offensive language and hate speech detection","author":"Vargas","year":"2021","journal-title":"arXiv:2103.14972"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W19-3510"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/BRACIS.2019.00084"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.5753\/jbcs.2025.5797"},{"key":"ref26","article-title":"LLaMA: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv:2302.13971"},{"key":"ref27","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023","journal-title":"arXiv:2307.09288"},{"key":"ref28","article-title":"The llama 3 herd of models","author":"Grattafiori","year":"2024","journal-title":"arXiv:2407.21783"},{"key":"ref29","article-title":"Qwen technical report","volume-title":"arXiv:2309.16609","author":"Bai","year":"2023"},{"key":"ref30","article-title":"Qwen2.5 technical report","volume-title":"arXiv:2412.15115","author":"Yang","year":"2024"},{"key":"ref31","article-title":"Qwen2 technical report","volume-title":"arXiv:2407.10671","author":"Yang","year":"2024"},{"key":"ref32","article-title":"Qwen3 technical report","volume-title":"arXiv:2505.09388","author":"Yang","year":"2025"},{"key":"ref33","volume-title":"The Falcon 3 Family of Open Models","year":"2024"},{"key":"ref34","article-title":"DeepSeek LLM: Scaling open-source language models with longtermism","author":"Bi","year":"2024","journal-title":"arXiv:2401.02954"},{"key":"ref35","first-page":"1","article-title":"Character-level convolutional networks for text classification","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Zhang"},{"key":"ref36","article-title":"Evaluating GPT-3.5 and GPT-4 models on Brazilian university admission exams","author":"Nunes","year":"2023","journal-title":"arXiv:2303.17003"},{"key":"ref37","first-page":"142","article-title":"Learning word vectors for sentiment analysis","volume-title":"Proc. 49th Annu. Meeting Assoc. Comput. Linguistics","author":"Maas"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.235"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00433"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D13-1170"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.5753\/eniac.2019.9334"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00370"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.naacl-main.52"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-naacl.149"},{"key":"ref45","first-page":"90","article-title":"Choice of plausible alternatives: An evaluation of commonsense causal reasoning","volume-title":"Proc. AAAI spring symposium: Log. formalizations commonsense reasoning","author":"Roemmele"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2020\/501"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/d19-1454"},{"key":"ref48","article-title":"Think you have solved question answering? Try ARC, the AI2 reasoning challenge","author":"Clark","year":"2018","journal-title":"arXiv:1803.05457"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/n16-1098"},{"key":"ref50","first-page":"1","article-title":"Measuring mathematical problem solving with the MATH dataset","volume-title":"Proc. NeurIPS","author":"Hendrycks"},{"key":"ref51","article-title":"Multiple-choice questions are efficient and robust LLM evaluators","author":"Zhang","year":"2024","journal-title":"arXiv:2405.11966"},{"key":"ref52","article-title":"Aligning AI with shared human values","author":"Hendrycks","year":"2020","journal-title":"arXiv:2008.02275"},{"key":"ref53","first-page":"9050","article-title":"InferBR: A natural language inference dataset in Portuguese","volume-title":"Proc. Joint Int. Conf. Comput. Linguistics, Lang. Resour. Eval.","author":"Bencke"},{"key":"ref54","first-page":"432","article-title":"RePro: A benchmark for opinion mining for Brazilian Portuguese","volume-title":"Proc. 16th Int. Conf. Comput. Process. Portuguese","volume":"1","author":"Santos Silva"},{"issue":"3","key":"ref55","first-page":"1","article-title":"Detec\u00e7\u00e3o de Discurso de \u00d3dio contra Mulheres em Textos em Portugu\u00eas Brasileiro: Constru\u00e7\u00e3o da Base MINA-BR e Modelo de Classifica\u00e7\u00e3o","volume":"20","author":"Plath","year":"2022","journal-title":"Revista Eletr\u00f4nica de Inicia\u00e7\u00e3o Cient\u00edfica em Computa\u00e7\u00e3o"}],"container-title":["IEEE Access"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/6287639\/10820123\/11303664.pdf?arnumber=11303664","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,26]],"date-time":"2025-12-26T05:35:43Z","timestamp":1766727343000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11303664\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025]]},"references-count":55,"URL":"https:\/\/doi.org\/10.1109\/access.2025.3645771","relation":{},"ISSN":["2169-3536"],"issn-type":[{"value":"2169-3536","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025]]}}}