ELITE: Experiential Learning and Intent-Aware Transfer for Self-improving Embodied Agents

TL;DR

ELITE is a framework that enables embodied agents to learn from their environment and transfer knowledge to similar tasks, significantly improving their performance in complex, real-world scenarios.

Contribution

The paper introduces ELITE, a novel embodied agent framework that combines experiential learning and intent-aware transfer to enhance task execution and generalization.

Findings

ELITE improves performance by 9% on EB-ALFRED benchmark.

ELITE achieves 5% higher success rate on EB-Habitat benchmark.

ELITE generalizes well to unseen task categories.

Abstract

Vision-language models (VLMs) have shown remarkable general capabilities, yet embodied agents built on them fail at complex tasks, often skipping critical steps, proposing invalid actions, and repeating mistakes. These failures arise from a fundamental gap between the static training data of VLMs and the physical interaction for embodied tasks. VLMs can learn rich semantic knowledge from static data but lack the ability to interact with the world. To address this issue, we introduce ELITE, an embodied agent framework with {E}xperiential {L}earning and {I}ntent-aware {T}ransfer that enables agents to continuously learn from their own environment interaction experiences, and transfer acquired knowledge to procedurally similar tasks. ELITE operates through two synergistic mechanisms, \textit{i.e.,} self-reflective knowledge construction and intent-aware retrieval. Specifically, self-reflective knowledge construction extracts reusable strategies from execution trajectories and maintains an evolving strategy pool through structured refinement operations. Then, intent-aware retrieval identifies relevant strategies from the pool and applies them to current tasks. Experiments on the EB-ALFRED and EB-Habitat benchmarks show that ELITE achieves 9\% and 5\% performance improvement over base VLMs in the online setting without any supervision. In the supervised setting, ELITE generalizes effectively to unseen task categories, achieving better performance compared to state-of-the-art training-based methods. These results demonstrate the effectiveness of ELITE for bridging the gap between semantic understanding and reliable action execution.