Results 11 to 20 of about 1,608,065 (365)
A dataset and exploration of models for understanding video data through fill-in-the-blank question-answering [PDF]
Computer Vision and Pattern Recognition, 2016 While deep convolutional neural networks frequently approach or exceed human-level performance in benchmark tasks involving static images, extending this success to moving images is not straightforward.
Tegan Maharaj +4 more
openalex +3 more sources
GLM: General Language Model Pretraining with Autoregressive Blank Infilling [PDF]
Annual Meeting of the Association for Computational Linguistics, 2021 There have been various types of pretraining architectures including autoencoding models (e.g., BERT), autoregressive models (e.g., GPT), and encoder-decoder models (e.g., T5).
Zhengxiao Du +6 more
semanticscholar +1 more source
Deep Search for Decaying Dark Matter with XMM-Newton Blank-Sky Observations. [PDF]
Physical Review Letters, 2021 Sterile neutrinos with masses in the keV range are well-motivated extensions to the Standard Model that could explain the observed neutrino masses while also making up the dark matter (DM) of the universe. If sterile neutrinos are DM then they may slowly
J. Foster +6 more
semanticscholar +1 more source
Multi-Blank Transducers for Speech Recognition [PDF]
IEEE International Conference on Acoustics, Speech, and Signal Processing, 2022 This paper proposes a modification to RNN-Transducer (RNN-T) models for automatic speech recognition (ASR). In standard RNN-T, the emission of a blank symbol consumes exactly one input frame; in our proposed method, we introduce additional blank symbols,
Hainan Xu +4 more
semanticscholar +1 more source
Microplastics analytics: why we should not underestimate the importance of blank controls
bioRxiv, 2023 Recent years have seen considerable scientific attention devoted towards documenting the presence of microplastics (MPs) in environmental samples. Due to omnipresence of environmental microplastics, however, disentangling environmental MPs from sample ...
M. Noonan +3 more
semanticscholar +1 more source
Conference on Empirical Methods in Natural Language Processing, 2020 We propose Blank Language Model (BLM), a model that generates sequences by dynamically creating and filling in blanks. The blanks control which part of the sequence to expand, making BLM ideal for a variety of text editing and rewriting tasks.
T. Shen +3 more
semanticscholar +1 more source
RFBFN: A Relation-First Blank Filling Network for Joint Relational Triple Extraction
Annual Meeting of the Association for Computational Linguistics, 2022 Joint relational triple extraction from unstructured text is an important task in information extraction. However, most existing works either ignore the semantic information of relations or predict subjects and objects sequentially. To address the issues,
Zhe Li +4 more
semanticscholar +1 more source
Earing Reduction by Varying Blank Holding Force in Deep Drawing with Deep Neural Network
Metals, 2021 In the present study, we propose a novel method of varying blank holding force (BHF) with the segmental blank holder and investigated its influence on the earing reduction in the circular deep drawing process of an aluminum alloy sheet.
M. Tran, Z. Shan, H. Lee, Dong-Kyu Kim
semanticscholar +1 more source