Learning to Learn from Multimodal Experience

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Sui, Xingyu, Zhao, Weixiang, Tang, Yongxin, Zhao, Yanyan, Wu, Yang, Tu, Dandan, Qin, Bing
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909049712279552
author Sui, Xingyu
Zhao, Weixiang
Tang, Yongxin
Zhao, Yanyan
Wu, Yang
Tu, Dandan
Qin, Bing
author_facet Sui, Xingyu
Zhao, Weixiang
Tang, Yongxin
Zhao, Yanyan
Wu, Yang
Tu, Dandan
Qin, Bing
contents Experience-driven learning has emerged as a promising paradigm for enabling agents to improve from interaction trajectories by accumulating and reusing past experience. However, existing approaches are predominantly developed in textual settings and rely on manually designed memory schemas, limiting their applicability to multimodal environments. In real-world scenarios, experience is inherently multimodal, involving heterogeneous signals across perception, reasoning, and action, which makes effective memory design significantly more challenging. In particular, the optimal way to structure and utilize multimodal experience is highly task-dependent and evolves over time, rendering fixed memory designs insufficient. In this work, we propose a new paradigm, learning to learn from multimodal experience, which shifts memory design from a predefined component to an adaptive and learnable process. Our framework enables agents to dynamically construct, organize, and utilize memory based on task requirements and interaction history, effectively learning how to structure experience for improved performance. Experiments demonstrate that adaptive memory design substantially enhances agent performance and generalization across multimodal tasks, highlighting the critical role of learning memory mechanisms in experience-driven learning.
format Preprint
id arxiv_https___arxiv_org_abs_2605_16857
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Learning to Learn from Multimodal Experience
Sui, Xingyu
Zhao, Weixiang
Tang, Yongxin
Zhao, Yanyan
Wu, Yang
Tu, Dandan
Qin, Bing
Artificial Intelligence
Experience-driven learning has emerged as a promising paradigm for enabling agents to improve from interaction trajectories by accumulating and reusing past experience. However, existing approaches are predominantly developed in textual settings and rely on manually designed memory schemas, limiting their applicability to multimodal environments. In real-world scenarios, experience is inherently multimodal, involving heterogeneous signals across perception, reasoning, and action, which makes effective memory design significantly more challenging. In particular, the optimal way to structure and utilize multimodal experience is highly task-dependent and evolves over time, rendering fixed memory designs insufficient. In this work, we propose a new paradigm, learning to learn from multimodal experience, which shifts memory design from a predefined component to an adaptive and learnable process. Our framework enables agents to dynamically construct, organize, and utilize memory based on task requirements and interaction history, effectively learning how to structure experience for improved performance. Experiments demonstrate that adaptive memory design substantially enhances agent performance and generalization across multimodal tasks, highlighting the critical role of learning memory mechanisms in experience-driven learning.
title Learning to Learn from Multimodal Experience
topic Artificial Intelligence
url https://arxiv.org/abs/2605.16857