Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Indonesian Journal of Electrical Engineering and Computer Science
Habib Hamam
Université de Moncton

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Exploring the intricacies of human memory and its analogous representation in ChatGPT Habib Hamam
Indonesian Journal of Electrical Engineering and Computer Science Vol 34, No 3: June 2024
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v34.i3.pp1760-1769

Abstract

Human memory and ChatGPT both rely on associations and patterns to generate contextually relevant responses. We explore how they work in tandem. Both use associations to activate related information when prompted. Memory forms generic representations that become precise with added details, similar to ChatGPT's responses with specific prompts. Activation Through Cues: Memory and ChatGPT recall based on cues or prompts, influenced by input. Level of Detail: Memory constructs mental images based on information, just as ChatGPT responds to input details. Dynamic Nature: Both adapt to memorize repeated segments with diverse continuations. By understanding the dynamics of memory and its parallels with ChatGPT's response generation, researchers can further enhance the model's capabilities. Fine-tuning the model's ability to activate relevant information, generate specific responses, and adapt to varying levels of detail and specificity in the input can contribute to its overall performance and relevance in various language tasks.
Co-Authors