LangSuitE: Planning, Controlling and Interacting with Large Language Models in Embodied Text Environments
This work addresses the need for better evaluation of LLMs as embodied agents for researchers in AI and robotics, though it is incremental as it builds on existing testbed approaches.
The paper tackles the problem of evaluating large language models as embodied agents in dynamic interactive environments by introducing LangSuitE, a versatile testbed with 6 embodied tasks, and finds that it reveals challenges in embodied planning through comprehensive benchmarks.
Recent advances in Large Language Models (LLMs) have shown inspiring achievements in constructing autonomous agents that rely on language descriptions as inputs. However, it remains unclear how well LLMs can function as few-shot or zero-shot embodied agents in dynamic interactive environments. To address this gap, we introduce LangSuitE, a versatile and simulation-free testbed featuring 6 representative embodied tasks in textual embodied worlds. Compared with previous LLM-based testbeds, LangSuitE (i) offers adaptability to diverse environments without multiple simulation engines, (ii) evaluates agents' capacity to develop ``internalized world knowledge'' with embodied observations, and (iii) allows easy customization of communication and action strategies. To address the embodiment challenge, we devise a novel chain-of-thought (CoT) schema, EmMem, which summarizes embodied states w.r.t. history information. Comprehensive benchmark results illustrate challenges and insights of embodied planning. LangSuitE represents a significant step toward building embodied generalists in the context of language models.