Problem
Research questions and friction points this paper is trying to address.
embodied intelligent virtual agents
cognitive architecture
Small Language Models
edge computing
Metaverse
Enhancing Virtual Agents through SLMs and Edge-Computing: An Exploratory Evaluation of Think and Memory Processes
🤖 AI Summary
This work addresses the pressing need for embodied intelligent agents in complex virtual and metaverse environments to exhibit persistent, adaptive, and context-aware cognitive capabilities. To this end, the authors propose a lightweight edge implementation of a Cognitive Embodied Agent Architecture (CEAA), which, for the first time, integrates small language models (SLMs)—specifically the Qwen2.5 series—onto an NVIDIA Jetson Orin NX platform to form a compact cognitive “brain” endowed with perception, memory, reasoning, and action faculties. Experimental results demonstrate that the system achieves strong performance in service request handling, memory-augmented dialogue, routing accuracy, and response latency, thereby validating that SLM-driven CEAA effectively supports cognitive persistence and efficient interaction for virtual agents operating at the edge.
📝 Abstract
Embodied intelligent virtual agents are expected to operate as persistent, adaptive, and context-aware entities within complex virtual and Metaverse worlds. However, implementing cognitively capable agents in such environments is conceptually and technologically challenging. Among a range of blueprints and development approaches, the Cognitive Embodied Agent Architecture (CEAA) has been developed as an implementation-oriented framework for architecting components of perception, memory, reasoning, planning, and embodied action. Considering the recent advances in edge computing and generative AI language models, this paper explores the use of Small Language Models (SLMs) to support edge-based operation of selected CEAA components, focusing on "Think" and "Memory" as processes central to cognitive orchestration and persistence of virtual agents in interactive virtual worlds. An edge-based virtual agent gateway system was developed and evaluated on an NVIDIA Jetson Orin NX using Qwen2.5 models of different sizes, exploring the system's capability to process service requests and handle memory-driven conversations. A series of simulation experiments evaluated routing accuracy, memory-read performance, and latency, demonstrating an SLM-driven prototype agent system that partially implements selected CEAA processes to support the development of embodied agents whose cognitive "brain" can operate efficiently and contextually for interactive experiences in immersive virtual worlds.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Small Language Models
Edge Computing
Cognitive Embodied Agent Architecture
Memory-Driven Interaction
Virtual Agents
🔎 Similar Papers
No similar papers found.
