Unverified50% confidenceFactExact time
Deformable convolution was proposed by Microsoft Research Asia in 2017 and adaptively focuses on irregularly shaped target regions by learning offsets for each sampling position of the convolution kernel.
1
Sources
50%
Confidence
Long-term
Relevance
9/4/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedDeformable DETR was proposed by Zhu Xizhou's team at Tsinghua University in 2021, and by concentrating attention on a small number of key sampling points through a deformable attention mechanism, it improves convergence speed by approximately 10 times.56% similarVerifiedA 2023 Stanford paper argued that emergence is a measurement artifact caused by the choice of evaluation metrics, and that with smooth metrics the capability improvement is continuous.56% similarUnverified掩码自编码器（MAE）由 He et al.（2021）提出，其核心创新在于采用 75% 极高遮盖率随机遮盖图像块后让模型重建56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/854227API
curl https://kongchang.com/api/v1/knowledge/claims/854227MCP
get_claim(id=854227)