相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Article
Computer Science, Artificial Intelligence
Yajie Bao et al.
Summary: Visual question answering requires processing multi-modal information and effective reasoning. Neural-symbolic learning is a promising method, but current approaches lack uncertainty handling and can only provide a single answer. To address this, we propose a confidence based neural-symbolic approach that evaluates NN inferences and conducts reasoning based on confidence.
Proceedings Paper
Computer Science, Artificial Intelligence
Zhou Yu et al.
Summary: Building benchmarks for VideoQA models is challenging yet crucial. Current benchmarks suffer from language biases, making it difficult to diagnose model weaknesses. We present ANetQA, a large-scale benchmark that supports fine-grained compositional reasoning over untrimmed videos. ANetQA is more fine-grained than existing benchmarks, and there is room for improvement according to experiments.
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2023)
Article
Computer Science, Artificial Intelligence
Huayi Zhan et al.
Summary: The method proposed in this paper utilizes key features such as entity-attribute graphs, query graphs, reinforcement learning models, and inference schemes to efficiently process visual tasks and accurately answer questions.
Proceedings Paper
Computer Science, Artificial Intelligence
Chenchen Jing et al.
Summary: This paper presents a dialog-like reasoning method to maintain reasoning consistency in answering a compositional question and its sub-questions. By integrating the reasoning processes for the sub-questions into the reasoning process for the compositional question like a dialog task, and using a consistency constraint to penalize inconsistent answer predictions, the effectiveness of the method is demonstrated through experimental results.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022)
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Zhuowan Li et al.
Summary: Neural symbolic methods demonstrate strong performance in synthetic images but struggle in real images, mainly due to the long-tail distribution of visual concepts and unequal importance of reasoning steps. The proposed CCO paradigm addresses these challenges by enabling models to capture underlying data characteristics and reason with hierarchical importance, significantly boosting their performance on real images and reducing the performance gap between symbolic and non-symbolic methods.
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Corentin Kervadec et al.
Summary: In this paper, it is argued that uncertainty in vision is a significant factor preventing successful learning of reasoning in vision and language problems. The study introduces a visual oracle that is less prone to exploiting spurious dataset biases, and proposes to transfer reasoning patterns from the oracle to a state-of-the-art Transformer-based VQA model. Experimental results show higher overall accuracy and accuracy on infrequent answers, indicating improved generalization and reduced dependency on dataset biases.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Wenhu Chen et al.
Summary: The Meta Module Network (MMN) addresses the scalability and generalizability issues of the Neural Module Network (NMN) by introducing a novel meta module and a flexible instantiation mechanism. MMN exhibits strong interpretability and compositionality in complex tasks, promising better scalability and generalizability.
2021 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV 2021)
(2021)
Article
Automation & Control Systems
Yajie Bao et al.
Summary: This letter presents a variational Bayesian inference Neural Network (BNN) approach to quantify uncertainties in matrix function estimation for the state-space linear parameter-varying (LPV) model identification problem using only inputs/outputs data. The proposed method simultaneously estimates states and posteriors of matrix functions given data.
IEEE CONTROL SYSTEMS LETTERS
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Ishan Misra et al.
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2018)
Proceedings Paper
Computer Science, Artificial Intelligence
Qingxing Cao et al.
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2018)
Proceedings Paper
Computer Science, Artificial Intelligence
Yuke Zhu et al.
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017)
(2017)
Proceedings Paper
Computer Science, Artificial Intelligence
Justin Johnson et al.
2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV)
(2017)
Proceedings Paper
Computer Science, Artificial Intelligence
Justin Johnson et al.
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017)
(2017)
Proceedings Paper
Computer Science, Artificial Intelligence
Kaiming He et al.
2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV)
(2017)
Proceedings Paper
Computer Science, Artificial Intelligence
Stanislaw Antol et al.
2015 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV)
(2015)
