Guilherme Silva
Guilherme Silva has 3 papers in our library, in 2026, most of them on Language models & pretraining, Sentiment & emotion, Multilingual & cross-lingual NLP; often with Matheus Peixoto, Pedro Silva, Eduardo Luz.
No OpenAlex profile matched yet.
Fields, topics and concepts
📚 Shelves 2The field of science this work belongs to (OpenAlex fields).Artificial Intelligence · 3 worksComputer Science · 3 works
🧭 Topics 4Research subjects and languages the work is about.Efficient models & compression · 1 workLanguage models & pretraining · 1 workMultilingual & cross-lingual NLP · 1 workSentiment & emotion · 1 work
Works by year
2020s
Papers
- A Multitask Transformer for Offensive Language Detection and Target Identification in HateBR (2026)
The paper argues that treating offensive language detection in Brazilian Portuguese as a single yes/no decision ignores the layered structure of toxicity. - Global vs. Local Sentence Embeddings for Brazilian Portuguese: Revisiting Monolingual Models in the Age of Foundation Models (2026)
The paper asks whether large multilingual ("global") foundation models, lightly adapted, can replace Portuguese-specialized ("local") models as sources of sentence embeddings for Brazilian Portuguese (PT-BR). - Lost in Quantization: Activation Outliers Explain Language-Specific FP8 Sensitivity in Llama-3 (2026)
Most evidence that 8-bit inference is nearly lossless comes from English evaluations, so it is unclear whether low-precision formats affect languages equally. The authors test Meta-Llama-3-8B on English and Brazilian Portuguese (PT-BR) under three regimes:
Co-authors here
Venues
- International Conference on Computational Processing of Portuguese ×3: International Conference on Computational Processing of Portuguese 2026 ×3
Updated
This page in Markdown → https://md.research.elamur.ai/free/en/authors/guilherme-silva/