研究人员正通过几种新方法推进视觉问答(VQA)。一篇论文介绍了VT-Transformer,它使用Transformer架构通过分析视觉和文本特征来预测答案的可预测性,在VizWiz 2020数据集上显示出有效性。另一项研究提出了一个框架,增强反事实对比学习以提高VQA模型对语言偏差的鲁棒性,在VQA-CP v2和VQA v2数据集上取得了强劲的性能。第三篇论文通过引入面向视觉的大型视频模型的基准来挑战当前的评估方法,突出了它们在视觉推理方面相对于基于文本的查询的弱点。最后,提出了一个名为USPLIT-VQA的拆分学习框架,用于隐私保护的VQA,并结合了贡献感知加权聚合机制来减轻恶意更新并减少内存和通信开销。
AI
arXiv:2609.19158v1 Announce Type: new Abstract: Knowledge graphs are usually integrated into question answering by encoding a retrieved subgraph with a graph neural network and fusing it with the language model in the online inference path. The same subgraph is therefore re-encod…
Answer grounding in document visual question answering remains an open challenge: most benchmarks lack grounding annotations or provide limited-quality labels, while constructing grounded datasets still requires costly manual effort. We introduce DocAttriBench (DAB), a large-scal…
arXiv cs.AI
TIER_1English(EN)·Tung Le, Huy Tien Nguyen, Le Minh Nguyen·
arXiv:2609.16565v1 Announce Type: cross Abstract: Answerability on Visual Question Answering is a novel and attractive task to predict answerable scores between images and questions in multi-modal data. Existing works often utilize a binary mapping from visual question answering …
arXiv cs.CV
TIER_1Deutsch(DE)·Luca De Grandis (University of Modena and Reggio Emilia, Modena, Italy), Silvia Cappelletti (University of Modena and Reggio Emilia, Modena, Italy), William Raccagni (University of Modena and Reggio Emilia, Modena, Italy, University of Pisa, Pisa, Italy)…·
arXiv:2609.20574v1 Announce Type: new Abstract: Answer grounding in document visual question answering remains an open challenge: most benchmarks lack grounding annotations or provide limited-quality labels, while constructing grounded datasets still requires costly manual effort…
arXiv cs.CV
TIER_1English(EN)·Rwiddhi Chakraborty (Oliver), Yinong (Oliver), Wang, Cheng Zhang, Fan Bai, Zhuoran You, Michael Kampffmeyer, Yong Jae Lee, Fernando De la Torre, Robert Jenssen·
arXiv:2609.17112v1 Announce Type: new Abstract: Large video models have exhibited impressive performance on a wide range of visual question answering tasks, owing to the rise of powerful, pretrained text and vision encoders. The usefulness of such models have also been demonstrat…
arXiv:2609.16567v1 Announce Type: new Abstract: Modern Visual Question Answering (VQA) models often exploit spurious correlations in training data, leading to poor out-of-distribution (OOD) generalization due to language bias. Although counterfactual learning has shown promise, e…
arXiv:2609.12168v1 Announce Type: new Abstract: Visual Question Answering (VQA) systems, jointly interpreting images and natural language queries, hold significant promise across many domains, yet the privacy-sensitive nature of user data creates a fundamental barrier. Centralize…