☆ Yσɠƚԋσʂ ☆@lemmygrad.ml to

Technology@lemmygrad.mlEnglish · 7 months ago

DeepSeek unveils new AI reasoning method amid anticipation for R2 model

2

cross-posted to:
[email protected]

28

DeepSeek unveils new AI reasoning method amid anticipation for R2 model

☆ Yσɠƚԋσʂ ☆@lemmygrad.ml to

Technology@lemmygrad.mlEnglish · 7 months ago

2

cross-posted to:
[email protected]

In collaboration with Tsinghua University, DeepSeek developed a technique combining reasoning methods to guide AI models towards human preferences.

You must log in or # to comment.

Chat

KrasnaiaZvezda@lemmygrad.ml
link
fedilink
arrow-up
5·
7 months ago
Link to the paper.
RedWizard [he/him, comrade/them]@hexbear.net
link
fedilink
English
arrow-up
5·
7 months ago
I feel like they just launched their reasoning model too, incredible.