Constraint Satisfaction
Constraint satisfaction problems (CSPs) involve finding assignments to variables that satisfy a set of constraints. Current research focuses on integrating constraint programming (CP) with machine learning models, particularly large language models (LLMs), to leverage the strengths of both approaches for improved efficiency and accuracy in solving complex CSPs, including those arising in text generation, robotics, and scheduling. This hybrid approach addresses limitations of LLMs in handling structural constraints and CP's difficulties with semantic understanding. The resulting advancements have significant implications for various fields requiring automated reasoning and decision-making under constraints.
27papers
Papers
February 18, 2025
December 3, 2024
November 21, 2024
October 31, 2024
October 28, 2024
October 14, 2024
July 16, 2024