Paulo Camara
Titles as the sources gave them. In 1 entry, with 0 checked quotations.
In the Base
Every entry they appear in.
Inherited Weights: Measuring the Tokenization Premium and the Sovereignty Cost of Language Models for the Global South
According to the abstract, Paulo Camara argues that fine-tuned models inherit vocabulary, knowledge, cultural alignment and licence from their base model. The paper measures how much more text costs in tokens across 31 languages and offers an eight-test audit a buyer can run without access to weights.