Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
Ukv
10 months ago
|
parent
|
context
|
favorite
| on:
UCSD: Large Language Models Pass the Turing Test
Human win rate would be 1 minus the model win rate, to my understanding. So 77% against ELIZA, 27% against GPT-4.5 with a human persona.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
