Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
cowsandmilk
on Dec 17, 2024
|
parent
|
context
|
favorite
| on:
New LLM optimization technique slashes memory cost...
More of lossy compression
aziaziazi
on Dec 17, 2024
|
next
[–]
Does lossy means you may see previous inputs change ?!
TkTech
on Dec 17, 2024
|
parent
|
next
[–]
Lossy doesn't necessarily imply that it is non-deterministic, just irreversible.
lxgr
on Dec 17, 2024
|
parent
|
prev
|
next
[–]
Only the model’s view, doesn’t have to be yours, just like you can participate in a long conversation without perfect memory that might in retrospect slightly differ from a recording.
verdverm
on Dec 17, 2024
|
prev
[–]
with the added bonus that these things are unreliable and the compressor could drop important tokens
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: