PEAR
📚
2
PEAR :Pixel-aligned Expressive humAn mesh Recovery
Liam Schoneveld
nlml
AI & ML interests
None yet
Recent Activity
liked
a Space
20 days ago
BestWJH/PEAR
updated
a Space
3 months ago
nlml/sheap
published
a Space
3 months ago
nlml/sheap
Organizations
None yet
liked
a
Space
20 days ago
updated
a
Space
3 months ago
published
a
Space
3 months ago
commented on
SmolLM3: smol, multilingual, long-context reasoner
8 months ago
Ah I see, thanks a lot!
commented on
SmolLM3: smol, multilingual, long-context reasoner
8 months ago
Great work! I have a question/curiosity. What is the advantage of splitting the model across two GPUs during training? (mentioned you use Tensor Parallelism (TP) of 2, Data Parallelism (DP) of 4 on each 8-GPU node). I am guessing the model can fit on a single GPU given it is small? I would have thought in such a case the most efficient implementation would be to use DP=8 ?
commented
a
paper
10 months ago
Control lips, teeth and tongue for lip sync task.
2
#20 opened almost 3 years ago
by
Temir