cd/entity/VQA-CP v2· home› entities› VQA-CP v2
grep -l @vqa-cp v2 /news/*.json | wc -l → 1

VQA-CP v2

mentions 1 type Person feed RSS

// recent coverage 1 mentions

04:00
2026-09-16
arxiv.org
computer-vision

Counterfactual Reasoning for Robust Visual Question Answering

A new training framework for Visual Question Answering (VQA) achieves 61.64% accuracy on the bias-sensitive VQA-CP v2 benchmark while maintaining 62.80% on the standard VQA v2 dataset, a generalizatio…

// co-occurs with top 5 entities
VQA v2 1 arXiv 1 Batch-Contrastive loss 1 Answer-Contrastive loss 1 Gradient-Discrepancy loss 1
// topics top 4 topics
computer vision 1 natural language processing 1 machine learning 1 ai research 1