AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution

2026-03-01

AI Summary

Current large language models struggle to translate user interaction experiences into reusable personalized capabilities, hindering the continuous accumulation of cross-session knowledge. This work proposes an experience-driven lifelong learning framework that, for the first time, explicitly abstracts interactive experiences into standardized, shareable, and composable skill units. It introduces a model-agnostic plugin architecture enabling automatic skill extraction, representation, retrieval, and dynamic injection. The framework facilitates continual skill evolution and cross-user transfer without requiring model retraining, effectively transforming ephemeral dialogue experiences into persistent, reusable personalized competencies. This approach provides a practical pathway toward building lifelong-learning agents and digital avatars capable of sustained personalization and adaptation.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Natural Language Processing: (Large) Language Models
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search

Abstract

In practical LLM applications, users repeatedly express stable preferences and requirements, such as reducing hallucinations, following institutional writing conventions, or avoiding overly technical wording, yet such interaction experience is seldom consolidated into reusable knowledge. Consequently, LLM agents often fail to accumulate personalized capabilities across sessions. We present AutoSkill, an experience-driven lifelong learning framework that enables LLM agents to automatically derive, maintain, and reuse skills from dialogue and interaction traces. AutoSkill abstracts skills from user experience, supports their continual self-evolution, and dynamically injects relevant skills into future requests without retraining the underlying model. Designed as a model-agnostic plugin layer, it is compatible with existing LLMs and introduces a standardized skill representation for sharing and transfer across agents, users, and tasks. In this way, AutoSkill turns ephemeral interaction experience into explicit, reusable, and composable capabilities. This paper describes the motivation, architecture, skill lifecycle, and implementation of AutoSkill, and positions it with respect to prior work on memory, retrieval, personalization, and agentic systems. AutoSkill highlights a practical and scalable path toward lifelong personalized agents and personal digital surrogates.

Problem

• lifelong learning
• personalization
• skill accumulation
• user experience
• large language models

Innovation

• lifelong learning
• skill self-evolution
• experience-driven
• model-agnostic
• personalized agents