Natural Language Constraint
Natural language constraints focus on improving the controllability and reliability of large language models (LLMs) by incorporating constraints expressed in natural language. Current research investigates how to effectively integrate these constraints into various tasks, including text generation, code generation, and multi-modal applications, often employing techniques like constrained decoding, constraint programming, and the integration of pre-trained language models. This research is significant because it addresses the limitations of LLMs in producing structured, accurate, and safe outputs, paving the way for more reliable and trustworthy applications in diverse fields.
Papers
October 5, 2024
July 3, 2024
June 24, 2024
June 18, 2024
June 15, 2024
June 6, 2024
May 31, 2024
May 30, 2024
May 27, 2024
May 8, 2024
February 7, 2024
January 15, 2024
October 10, 2023
December 20, 2022
July 5, 2022