註冊並分享邀請連結,可獲得影片播放與邀請獎勵。

ρ:ɡeon
@pigeon__s
omnimodality is all you need
加入 October 2020
69 正在關注    2K 粉絲
DeepSeek is too focused on architectural and memory usage efficiency and not really caring about changing up how their models reason or the reasoning efficiency their models still burn tokens like no other models on earth I feel like the RL they use is still pretty primitive while the memory improvements like KV cache and stuff for hyperscaling has been the focus
顯示更多
0
19
126
2
轉發到社區