Search papers, labs, and topics across Lattice.
This paper reviews the integration of large language models (LLMs), structured knowledge bases (KBs), and reasoning capabilities to advance general embodied intelligence (GEI). It highlights the evolution of LLM-centered systems and proposes a conceptual framework that illustrates the synergy among LLMs, KBs, reasoning, and physical embodiment, while identifying five key challenges that must be addressed for effective deployment. The findings underscore the necessity for adaptive, multimodal agents that can learn and act in complex environments, paving the way for future advancements in AI agents.
The integration of LLMs, knowledge bases, and reasoning capabilities could redefine how AI agents learn and operate in dynamic physical environments.
The convergence of large language models (LLMs), structured knowledge bases (KBs), and reasoning ability (RA) presents a promising trajectory toward general embodied intelligence (GEI). This paper reviews the evolution of LLM-centered intelligent systems, emphasising their integration with knowledge representation, logical reasoning, and physical embodiment. We analyse LLM architectures, pre-training methods, and inference mechanisms, along with their interaction with external knowledge sources and structured reasoning frameworks. Furthermore, we examine embodied intelligence (EI) paradigms wherein agents learn and act in physical environments. To synthesise these dimensions, we present a conceptual framework that illustrates the synergy among LLMs, KBs, RA, and embodiment, serving as a guiding model for perception, reasoning, and action rather than an implemented engineering architecture. To advance toward GEI, we identify five key challenges: efficient LLM deployment, closed-loop knowledge integration, hybrid symbolic-neural reasoning, perception-action grounding, and continual learning. This survey provides a comprehensive roadmap for developing adaptive, multimodal agents capable of operating in complex, dynamic settings.
