ExpeTrans: LLMs Are Experiential Transfer Learners

TL;DR

ExpeTrans introduces an autonomous framework enabling large language models to transfer task-solving experience across different tasks, reducing reliance on human-labeled data and enhancing generalization capabilities.

Contribution

The paper presents a novel autonomous experience transfer framework for LLMs, allowing them to mimic human-like transfer learning without extensive human effort.

Findings

Improves LLM performance on 13 datasets

Reduces need for human-labeled experience data

Enhances generalization to new tasks

Abstract

Recent studies provide large language models (LLMs) with textual task-solving experiences via prompts to improve their performance. However, previous methods rely on substantial human labor or time to gather such experiences for each task, which is impractical given the growing variety of task types in user queries to LLMs. To address this issue, we design an autonomous experience transfer framework to explore whether LLMs can mimic human cognitive intelligence to autonomously transfer experience from existing source tasks to newly encountered target tasks. This not only allows the acquisition of experience without extensive costs of previous methods, but also offers a novel path for the generalization of LLMs. Experimental results on 13 datasets demonstrate that our framework effectively improves the performance of LLMs. Furthermore, we provide a detailed analysis of each module in the framework.