TL;DR
CoEx introduces a hierarchical, co-evolving world model and exploration strategy for LLM-based agents, enabling dynamic updates and improved planning in complex environments.
Contribution
It presents a novel hierarchical architecture with a neurosymbolic belief state that co-evolves with the world model, enhancing planning and exploration capabilities.
Findings
Outperforms existing agents in complex tasks
Effective dynamic updating of the world model
Improved planning and exploration results
Abstract
Planning in modern LLM agents relies on the utilization of LLM as an internal world model, acquired during pretraining. However, existing agent designs fail to effectively assimilate new observations into dynamic updates of the world model. This reliance on the LLM's static internal world model is progressively prone to misalignment with the underlying true state of the world, leading to the generation of divergent and erroneous plans. We introduce a hierarchical agent architecture, CoEx, in which hierarchical state abstraction allows LLM planning to co-evolve with a dynamically updated model of the world. CoEx plans and interacts with the world by using LLM reasoning to orchestrate dynamic plans consisting of subgoals, and its learning mechanism continuously incorporates these subgoal experiences into a persistent world model in the form of a neurosymbolic belief state, comprising…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
