Owner
SHI-Labs
2 indexed repositories · View on GitHub
-
NATTEN
Fast Multi-dimensional Sparse Attention
C++ · 797 stars
-
physical-ai-bench
[CVPR 2026 Oral] PAI-Bench: A Comprehensive Benchmark for Physical AI
Python · 99 stars
-
Difficulty 3/5 1-2 days Newbie friendliness 45/100
-
Int32 offset overflow in the compute-delta reduction silently corrupts dQ/dK for long sequences Open
Difficulty 4/5 3-5 days Newbie friendliness 55/100
-
Difficulty 2/5 1-3 hours Newbie friendliness 88/100
-
Difficulty 4/5 3-5 days Newbie friendliness 25/100
SHI-Labs/physical-ai-bench#9 · 1 reaction ·
-
Difficulty 4/5 3-5 days Newbie friendliness 45/100
-
Difficulty 4/5 3-5 days Newbie friendliness 45/100
-
Difficulty 3/5 1-2 days Newbie friendliness 35/100
SHI-Labs/physical-ai-bench#6 · 1 comment ·
-
Difficulty 2/5 1-3 hours Newbie friendliness 35/100
-
Flash Attention 4 Open
Difficulty 5/5 Over a week Newbie friendliness 25/100
-
Difficulty 4/5 3-5 days Newbie friendliness 35/100
-
Difficulty 5/5 Over a week Newbie friendliness 15/100
-
RoPE vs RPB Open
Difficulty 5/5 Over a week Newbie friendliness 35/100
-
feature-request
Difficulty 5/5 Over a week Newbie friendliness 30/100
-
Difficulty 4/5 3-5 days Newbie friendliness 25/100
-
feature-request
Difficulty 5/5 Over a week Newbie friendliness 30/100
-
feature-request
Difficulty 4/5 3-5 days Newbie friendliness 35/100
-
feature-request
Difficulty 4/5 3-5 days Newbie friendliness 42/100
-
Arbitrary masks Openfeature-request
Difficulty 5/5 Over a week Newbie friendliness 25/100
-
Nested tensors Openfeature-request
Difficulty 5/5 Over a week Newbie friendliness 25/100