Corpus Level

Corpus-level analysis examines large collections of text data to uncover patterns and insights not readily apparent at the individual sentence or document level. Current research focuses on leveraging these analyses to detect biases (e.g., in Wikipedia articles), improve machine translation evaluation by refining aggregation methods, and monitor the impact of large language models on text generation. This work is significant for advancing computational social science, improving natural language processing tasks like named entity recognition, and understanding the evolving relationship between humans and AI in information creation and dissemination.