HumanAL: Calibrating Human Matching Beyond a Single Task
원문 초록
This work offers a novel view on the use of human input as labels, acknowledging that humans may err. We build a behavioral profile for human annotators which is used as a feature representation of the provided input. We show that by utilizing black-box machine learning, we can take into account human behavior and calibrate their input to improve the labeling quality. To support our claims and provide a proof-of-concept, we experiment with three different matching tasks, namely, schema matching, entity matching and text matching. Our empirical evaluation suggests that the method can improve the quality of gathered labels in multiple settings including cross-domain (across different matching tasks).
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Efficient Medical Vision-Language Alignment Through Adapting Masked Vision Models
Aligning Information Capacity Between Vision and Language via Dense-to-Sparse Feature Distillation for Image-Text Matching
CLIP is Strong Enough to Fight Back: Test-time Counterattacks towards Zero-shot Adversarial Robustness of CLIP
IteRPrimE: Zero-shot Referring Image Segmentation with Iterative Grad-CAM Refinement and Primary Word Emphasis
CLIP Under the Microscope: A Fine-Grained Analysis of Multi-Object Representation
FiLo++: Zero-/Few-Shot Anomaly Detection by Fused Fine-Grained Descriptions and Deformable Localization
토론
이 논문과 연결된 포럼 글이 아직 없습니다.