AI RESEARCH
DIA-HARM: Dialectal Disparities in Harmful Content Detection Across 50 English Dialects
arXiv CS.CL
•
ArXi:2604.05318v1 Announce Type: new Harmful content detectors-particularly disinformation classifiers-are predominantly developed and evaluated on Standard American English (SAE), leaving their robustness to dialectal variation unexplored. We present DIA-HARM, the first benchmark for evaluating disinformation detection robustness across 50 English dialects spanning U. S., British, African, Caribbean, and Asia-Pacific varieties. Using Multi-VALUE's linguistically grounded transformations, we.