arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.03381cs.CLcs.AI
收藏

CogniDual框架:双系统理论框架下自训练大语言模型以提升认知任务表现

CogniDual Framework: Self-Training Large Language Models within a Dual-System Theoretical Framework for Improving Cognitive Tasks

School of Electronic and Electrical Engineering, Shanghai University of Engineering Science(上海工程技术大学电子与电气工程学院);INF Technology (shanghai) Co., Ltd.(INF科技(上海)有限公司);School of Art, Design and Architecture, Monash University(莫纳什大学艺术、设计与建筑学院);Artificial Intelligence Innovation and Incubation Institute, Fudan University(复旦大学人工智能创新与孵化研究院)

Yongxin Deng, Xihe Qiu, Xiaoyu Tan, Chao Qu, Jing Pan, Yuan Cheng, Yinghui Xu, Wei Chu

AI总结:

本研究提出CogniDual框架,基于双系统理论探究大语言模型能否通过自训练从审慎推演转向直觉响应,揭示其认知机制并提升推理效率。

摘要:

认知心理学研究感知、注意、记忆、语言、问题解决、决策与推理。Kahneman的双系统理论阐释了人类决策过程,区分了快速直觉的系统1和审慎理性的系统2。近期进展使大语言模型(LLMs)成为在各类认知任务中接近人类水平的强大工具,但LLMs中是否存在类似人类认知的双系统框架仍未被探索。本研究提出面向LLMs的CogniDual框架(CFLLMs),旨在评估LLMs能否通过自训练从审慎推演演进为直觉响应,从而模拟人类获取并掌握新信息的过程。研究结果揭示了LLMs生成响应背后的认知机制,深化了我们对其认知心理学层面能力的理解。在实践层面,自训练模型可对特定查询提供更快响应,降低推理阶段的计算需求。

英文摘要:

Cognitive psychology investigates perception, attention, memory, language, problem-solving, decision-making, and reasoning. Kahneman's dual-system theory elucidates the human decision-making process, distinguishing between the rapid, intuitive System 1 and the deliberative, rational System 2. Recent advancements have positioned large language Models (LLMs) as formidable tools nearing human-level proficiency in various cognitive tasks. Nonetheless, the presence of a dual-system framework analogous to human cognition in LLMs remains unexplored. This study introduces the \textbf{CogniDual Framework for LLMs} (CFLLMs), designed to assess whether LLMs can, through self-training, evolve from deliberate deduction to intuitive responses, thereby emulating the human process of acquiring and mastering new information. Our findings reveal the cognitive mechanisms behind LLMs' response generation, enhancing our understanding of their capabilities in cognitive psychology. Practically, self-trained models can provide faster responses to certain queries, reducing computational demands during inference.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑