Adding chain-of-thought prompts with hand-picked picture cues to an LLM classifier improves Alzheimer's detection accuracy on ADReSS from 75% to 83.3% with ASR transcripts and to 87.5% with manual transcripts, though the reported 16.7% gain mixes mismatched settings.
It significantly impacts patients’ ability to perform daily activities, thereby severely affecting their quality of life
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Reasoning-Based Approach with Chain-of-Thought for Alzheimer's Detection Using Speech and Large Language Models
Adding chain-of-thought prompts with hand-picked picture cues to an LLM classifier improves Alzheimer's detection accuracy on ADReSS from 75% to 83.3% with ASR transcripts and to 87.5% with manual transcripts, though the reported 16.7% gain mixes mismatched settings.