{"name":"llm-legislativo","repo":"BaruqueRodrigues/llm-legislativo","url":"https://midorreal.com/project/baruquerodrigues-llm-legislativo","repository":"https://github.com/BaruqueRodrigues/llm-legislativo","description":"Corpus legislativo brasileiro em larga escala e modelo Transformer causal treinado do zero em PyTorch puro.","facts":{"license":{"value":"MIT","source":"https://github.com/BaruqueRodrigues/llm-legislativo","checked":"22 Nov 2026"},"language":{"value":"Python","source":"https://github.com/BaruqueRodrigues/llm-legislativo","checked":"22 Nov 2026"},"releases":{"value":"0","source":"https://github.com/BaruqueRodrigues/llm-legislativo/releases","checked":"19 Nov 2026"},"contributors":{"value":"1","source":"https://github.com/BaruqueRodrigues/llm-legislativo/graphs/contributors","checked":"19 Nov 2026"},"archived":{"value":"no","source":"https://github.com/BaruqueRodrigues/llm-legislativo","checked":"22 Nov 2026"}},"stars":501.0,"writeup":{"what_it_is":"A large-scale corpus of Brazilian legislative documents from Congress (Chamber of Deputies and Federal Senate) paired with a causal Transformer language model trained from scratch in pure PyTorch. The project includes 360+ GiB of official documents, a custom byte-level BPE tokenizer with 8,192 tokens, and a 6-layer decoder model trained on 1.25 billion tokens.","audience":"Researchers in NLP, Brazilian legislative analysis, and open-source AI.","claims":[{"kind":"specific","claim":"Transformer decoder implemented entirely in pure PyTorch without pre-trained weights or closed APIs","excerpt":"implementado do zero sem uso de pesos pré-treinados, sem dependência de APIs fechadas","status":"not_checked"},{"kind":"specific","claim":"Uses strict causal isolation with zero data leakage between train and test splits via family-based partitioning","excerpt":"Nenhum documento de uma mesma família tramitada divide conjuntos de treino e teste","status":"not_checked"}],"alternatives":[],"written_by":"AI, from the project's README","date":"2026-10-15"},"why_now":null,"owner_supplied":null,"score":{"verdict":"mid","hype":16,"reality":5,"gap":11,"momentum":null,"confidence":60,"version":"score-2.0","rule":"verdict-2.0","calculated_at":"2026-11-22T08:00:00+00:00","verdict_source":"rule"},"early_read":null,"changes":[{"date":"2026-10-13T23:46:21+00:00","text":"Repository created","source":"https://github.com/BaruqueRodrigues/llm-legislativo"}],"cite":"llm-legislativo is a GitHub project at https://github.com/BaruqueRodrigues/llm-legislativo written mainly in Python, described there as \"Corpus legislativo brasileiro em larga escala e modelo Transformer causal treinado do zero em PyTorch puro\".","attribution":"Data from Mid or Real, https://midorreal.com/project/baruquerodrigues-llm-legislativo, checked 22 November 2026. Attribution with a link is required.","still_real":null}