LoTa-Bench: Benchmarking Language-oriented Task Planners for Embodied Agents
This provides a standardized evaluation tool for researchers in embodied AI, though it is incremental as it builds on existing datasets and simulators.
The authors tackled the difficulty in comparing language-oriented task planners for embodied agents by proposing LoTa-Bench, a benchmark system that automatically quantifies performance on home-service tasks using datasets like ALFRED and AI2-THOR, and found it accelerates development through extensive experiments with LLMs and prompts.
Large language models (LLMs) have recently received considerable attention as alternative solutions for task planning. However, comparing the performance of language-oriented task planners becomes difficult, and there exists a dearth of detailed exploration regarding the effects of various factors such as pre-trained model selection and prompt construction. To address this, we propose a benchmark system for automatically quantifying performance of task planning for home-service embodied agents. Task planners are tested on two pairs of datasets and simulators: 1) ALFRED and AI2-THOR, 2) an extension of Watch-And-Help and VirtualHome. Using the proposed benchmark system, we perform extensive experiments with LLMs and prompts, and explore several enhancements of the baseline planner. We expect that the proposed benchmark tool would accelerate the development of language-oriented task planners.