.png){kind=icon}
.png){kind=icon}
Return
Neural feature alignment between large language models and brain activities: A knowledge-based framework for cross-modal analysis
.png){kind=icon}
PRE
AI
DOI：10.1016/j.neunet.2026.108916.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Knowledge engineering framework bridging AI and brain via interpretable features. • Neuroscience-grounded metrics predict model capabilities (r=0.736–0.886) reliably. • Multi-scale computational-cognitive correspondences reveal brain-like processing. • Training strategies sculpt targeted cognitive plausibility in knowledge systems. • Complex scaling patterns guide intelligent system design beyond simple expansion.
Keywords:
Neural feature alignment
Brain-like processing
Knowledge engineering
Cognitive plausibility
Cross-modal analysis
Journal
IF:
6.3
Papers:
7.8K
Citations:
3.0W