The field of planning and scheduling is witnessing a significant shift towards the adoption of Large Language Models (LLMs) to automate and improve various aspects of the process. Researchers are exploring ways to regulate LLMs to ensure reliable and precise constraint specification, which is a critical prerequisite for effective planning and scheduling. The integration of LLMs with formal methods and symbolic planning is also being investigated to enhance the reliability and repeatability of planning systems. Furthermore, the development of new action representations, such as planning with schemas, is being studied to improve the scalability and effectiveness of long-horizon planning agents. Noteworthy papers in this area include:

• Automated Constraint Specification for Job Scheduling by Regulating Generative Model with Domain-Specific Representation, which presents a constraint-centric architecture for reliable automated constraint specification.
• Constrained Natural Language Action Planning for Resilient Embodied Systems, which introduces a new robotic planning method that augments LLM planners with symbolic planning oversight to improve reliability and repeatability.
• The Cognitive Bandwidth Bottleneck: Shifting Long-Horizon Agent from Planning with Actions to Planning with Schemas, which systematically studies the effectiveness of different action representations for long-horizon planning agents.