Language User
Research on language users focuses on understanding how people interact with and utilize language technologies, particularly large language models (LLMs). Current investigations explore the limitations of LLMs in accurately representing diverse languages and cultures, including biases in information retrieval and the challenges of multilingual support. This research utilizes various model architectures, such as encoder-decoder transformers, to improve LLM performance and address issues like code-mixing in search queries. Ultimately, this work aims to create more equitable and effective language technologies that better serve the needs of diverse global populations.
Papers
May 23, 2024
February 1, 2024
July 26, 2023
May 20, 2023
August 7, 2022