Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
bitexploder
11 months ago
|
parent
|
context
|
favorite
| on:
LLMs don't do formal reasoning
Wonder if this is like the old school benchmarks people would cheat on. Should not be hard to assemble a series of such puzzles and get a read on overall accuracy :)
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: