GitHub

Hierarchical Prompt Learning Using CLIP for Multi-label Classification with Single Positive Labels. ACMMM 2023.

Collecting full annotations to construct multi-label datasets is difficult and labor-consuming. As an effective solution to relieve the annotation burden, single positive multi-label learning (SPML) ...
IEEE

Using Human Mobility Patterns to Forecast Outliers in Citizen Complaints Data

Abstract: Cities have been growing ever larger especially in the most recent decades. As neighborhoods and boundaries of cities are expanding, the resident-service related problems within these cities ...
NDTV

Rani Mukerji To NDTV On 'Women-Centric' Films: "It's A Label We Should Stop Using"

One of the key discussions was how films led by women are still categorised differently. Rani made it clear that she believes the label is outdated. "It's a label that we should stop using, and I ...
IEEE

Estimating Multi-Label Expected Accuracy Using Labelset Distributions

Abstract: A multi-label classifier estimates the binary label state (relevant/irrelevant) for each of a set of concept labels, for a given instance. Probabilistic multi-label classifiers provide a ...