AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

MPCViT: Searching for Accurate and Efficient MPC-Friendly Vision Transformer with Heterogeneous Attention

Wenxuan Zeng, Meng Li, Wenjie Xiong, Tong Tong, Wen-jie Lu, Jin Tan, Runsheng Wang, Ru Huang

ICCV 2023 · Architectures and efficiency › Attention and transformers · approach: Hybrid and interpretable ViT designs · 36 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace Softmax attention with ReLU Softmax (ablation)ViT attention in MPC · CIFAR-10top-1 accuracy (%) / latency (s): 6.82 s -> 5.32 s; 92.69 -> 92.31Improved
replace all Softmax with Scaling attention (ablation)ViT attention in MPC · CIFAR-10top-1 accuracy (%): 92.69 -> 91.52; 6.82 s -> 0.66 sMixed
use Linformer low-rank attention (ablation)ViT attention in MPC · CIFAR-10top-1 accuracy (%): 92.69 -> 90.85Hurt
use head-wise rather than layer-wise or row-wise heterogeneous attention search (ablation)MPCViT · CIFAR-10 (mu=0.5)top-1 accuracy (%): 93.01 / 93.16 -> 93.21Improved
add logits-based knowledge distillation (ablation)MPCViT · Tiny-ImageNet (mu=0.7)top-1 accuracy (%): 59.02 -> 62.39Improved
add feature-based KD on top of logits KD (ablation)MPCViT · Tiny-ImageNet (mu=0.7)top-1 accuracy (%): 62.39 -> 63.03Improved
add additional ReLU after fused linear layer in MPCViT+ (ablation)MPCViT+ · CIFAR-10 (mu=0.7)top-1 accuracy (%): 93.81 -> 94.27Improved
linearize GeLU in MLP (MPCViT+)MPCViT · CIFAR-10/100top-1 accuracy (%)Mixed
search per-head replacement instead of per-layer uniform replacement (ablation)MPCViT · CIFAR-10/100, Tiny-ImageNetaccuracy-latency Pareto frontImproved
replace attention heads with ScaleAttn instead of pruning heads (ablation)MPCViT · CIFAR-10top-1 accuracy (%): 93.03 -> 93.38Improved
mix Softmax attention with ScaleAttn instead of ReLU Softmax (ablation)MPCViT · CIFAR-10/100, Tiny-ImageNetlatency: at least 1.4x and 1.2x latencyHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy