提出LAPITHS框架，质疑大模型认知能力宣称的合理性
Taming the Centaur(s) with LAPITHS: a framework for a theoretically grounded interpretation of AI performances
- 基于理论认知网格与行为对比，构建可解释的评估框架
- 证明类似Centaurs模型的表现可由非认知系统复现
- 适合关注AI认知声称可信度的研究者参考
我们提出LAPITHS（通过范式基础的关于人类相似性的论点进行语言模型分析）框架，用以揭示如Centaurs等模型所宣称的统一认知人工系统中多项核心主张在理论与实证层面均缺乏依据。LAPITHS提供了一个严谨的参照点，以对抗当前主流研究将基于Transformer的语言模型达到人类水平表现，直接解读为具备类人底层计算机制乃至认知能力的趋势。其创新性在于引入两项量化评估：(i) 最小认知网格，一种基于理论推导的认知可行性估算方法；(ii) 行为对照实验，表明与Centaurs模型类似的结果，可在不具备典型认知可塑性结构且无法独立解释人类认知的其他系统中重现。
原文摘要 · Abstract (English)
We introduce a framework called LAPITHS (Language model Analysis through Paradigm grounded Interpretations of Theses about Human likenesS) and use it to show that several major claims advanced by models such as CENTAUR, proposed as an artificial Unified Model of Cognition, are not theoretically or empirically justified. LAPITHS provides a principled reference point for counteracting the current behaviouristic tendency in AI research to interpret the human level performances of transformer based language models as evidence of human like underlying computation and, by extension, as signs of cognitive abilities. The novelty of LAPITHS lies in making explicit the arguments grounded in two quantitative assessments: (i) the Minimal Cognitive Grid, a theoretically motivated method for estimating the cognitive plausibility of artificial systems, and (ii) a behavioural comparison showing that results similar to those reported for CENTAUR like models can be reproduced by other systems that do not satisfy the structural constraints typically associated with cognitive plausibility, and whose outputs do not provide independent explanatory insight into human cognition.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。