TANGO: オープンワールドタスクのための訓練不要な身体性AIエージェント
TANGO: Training-free Embodied AI Agents for Open-world Tasks
LLMに少数の例を与えて基本動作を組み合わせさせることで、追加学習なしに物体探索・生涯ナビゲーション・質問応答などの身体性AIタスクをゼロショットで解く手法を提案。
著者: Filippo Ziliotto, Tommaso Campari, Luciano Serafini, Lamberto Ballan
分類: cs.AI, cs.RO
原文アブストラクト
Large Language Models (LLMs) have demonstrated excellent capabilities in composing various modules together to create programs that can perform complex reasoning tasks on images. In this paper, we propose TANGO, an approach that extends the program composition via LLMs already observed for images, aiming to integrate those capabilities into embodied agents capable of observing and acting in the world. Specifically, by employing a simple PointGoal Navigation model combined with a memory-based exploration policy as a foundational primitive for guiding an agent through the world, we show how a single model can address diverse tasks without additional training. We task an LLM with composing the provided primitives to solve a specific task, using only a few in-context examples in the prompt. We evaluate our approach on three key Embodied AI tasks: Open-Set ObjectGoal Navigation, Multi-Modal Lifelong Navigation, and Open Embodied Question Answering, achieving state-of-the-art results without any specific fine-tuning in challenging zero-shot scenarios.
