Scribble Hides Class: Promoting Scribble-Based Weakly-Supervised Semantic Segmentation with Its Class Label
Scribble-based weakly-supervised semantic segmentation using sparse scribble
supervision is gaining traction as it reduces annotation costs when compared to fully …
supervision is gaining traction as it reduces annotation costs when compared to fully …
3M: An Affinity Processing Method for Weakly Supervised Semantic Segmentation
S Wang, N Cang - 2023 5th International Academic Exchange …, 2023 - ieeexplore.ieee.org
In the face of the high cost of annotation brought by fully supervised semantic segmentation,
a way to simplify the annotation has been proposed—weakly supervised semantic …
a way to simplify the annotation has been proposed—weakly supervised semantic …
[PDF][PDF] On the Robustness of In-Context Learning with Noisy Labels: Train, Inference, and Beyond
C Cheng, H Wen, X Yu, Z Wei - chencheng.me
Abstract Recently, the mysterious In-Context Learning (ICL) ability of Transformer
architecture, particularly in large language models, has garnered considerable research …
architecture, particularly in large language models, has garnered considerable research …