Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
red75prime
9 days ago
|
parent
|
context
|
favorite
| on:
MIT Technology Review has confirmed that posts on ...
It's relatively easy to analyze skill files. Shared chunks of neural networks (LLM adapters) can hide malicious behaviors better.
help
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
