New framework ranks cognitive plausibility of AI analogy and metaphor models

Researchers have developed a framework called the Minimal Cognitive Grid (MCG) to assess the cognitive plausibility of computational models. This framework was applied to evaluate prominent models of analogy and metaphor, such as the Structure-Mapping Engine (SME), CogSketch, METCL, and Large Language Models (LLMs). The analysis, based on functional/structural ratio, generality, and performance match, provides a quantitative comparison of these models against established cognitive theories. AI

IMPACT Introduces a new quantitative framework for evaluating the cognitive plausibility of AI models, potentially guiding future research in analogy and metaphor.

RANK_REASON Academic paper presenting a new framework for evaluating computational models. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• arXiv
• CogSketch
• Large Language Models
• LLMs
• Minimal Cognitive Grid
• SME
• Structure-Mapping Engine

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →