An investigation into the fairness and bias implications of multimodal fusion techniques in AI-driven recruitment systems using the FairCVdb dataset, examining how different fusion strategies influence bias and impact fairness in hiring decisions.
A dataset for improving automated fact-checking by addressing leaked and unreliable web-collected evidence.
A robust benchmark for multimodal misinformation detection, including out-of-context and miscaptioned images, that mitigates unimodal biases.
LESA: Linguistic Encapsulation and Semantic Amalgamation Based Generalised Claim Detection from Online Content
XLM-RoBERTa model which was trained during experiments in order to create models capable of detecting check-worthy claims in the widest achievable range.
mDeBERTa V3 model which was trained during experiments in order to create models capable of detecting check-worthy claims in the widest achievable range.