Results 11 to 20 of about 2,435,950 (372)
Information-Theoretic Probing for Linguistic Structure [PDF]
Annual Meeting of the Association for Computational Linguistics, 2020 The success of neural networks on a diverse set of NLP tasks has led researchers to question how much these networks actually “know” about natural language. Probes are a natural way of assessing this.
Tiago Pimentel +5 more
semanticscholar +1 more source
TNOA Journal of Ophthalmic Science and Research, 2022 A 3-year-old female child presented to us with complaints of watering in the left eye since birth, along with swelling near the medial canthus of the left eye for 1 week. External examination revealed a burst lacrimal sac abscess.
A Ajeeba Sheerin +2 more
doaj +1 more source
Probing Pretrained Language Models for Lexical Semantics [PDF]
Conference on Empirical Methods in Natural Language Processing, 2020 The success of large pretrained language models (LMs) such as BERT and RoBERTa has sparked interest in probing their representations, in order to unveil what types of knowledge they implicitly capture.
Ivan Vulic +4 more
semanticscholar +1 more source
Low-Complexity Probing via Finding Subnetworks [PDF]
North American Chapter of the Association for Computational Linguistics, 2021 The dominant approach in probing neural networks for linguistic properties is to train a new shallow multi-layer perceptron (MLP) on top of the model’s internal representations. This approach can detect properties encoded in the model, but at the cost of
Steven Cao +2 more
semanticscholar +1 more source
Perturbed Masking: Parameter-free Probing for Analyzing and Interpreting BERT [PDF]
Annual Meeting of the Association for Computational Linguistics, 2020 By introducing a small set of additional parameters, a probe learns to solve specific linguistic tasks (e.g., dependency parsing) in a supervised manner using feature representations (e.g., contextualized embeddings).
Zhiyong Wu, Yun Chen, B. Kao, Qun Liu
semanticscholar +1 more source
Learning Site-Specific Probing Beams for Fast mmWave Beam Alignment [PDF]
IEEE Transactions on Wireless Communications, 2021 Beam alignment – the process of finding an optimal directional beam pair – is a challenging procedure crucial to millimeter wave (mmWave) communication systems.
Yuqiang Heng, Jianhua Mo, J. Andrews
semanticscholar +1 more source
Conditional probing: measuring usable information beyond a baseline [PDF]
Conference on Empirical Methods in Natural Language Processing, 2021 Probing experiments investigate the extent to which neural representations make properties—like part-of-speech—predictable. One suggests that a representation encodes a property if probing that representation produces higher accuracy than probing a ...
John Hewitt +3 more
semanticscholar +1 more source
The What-If Tool: Interactive Probing of Machine Learning Models [PDF]
IEEE Transactions on Visualization and Computer Graphics, 2019 A key challenge in developing and deploying Machine Learning (ML) systems is understanding their performance across a wide range of inputs. To address this challenge, we created the What-If Tool, an open-source application that allows practitioners to ...
James Wexler +5 more
semanticscholar +1 more source
Discourse Probing of Pretrained Language Models [PDF]
North American Chapter of the Association for Computational Linguistics, 2021 Existing work on probing of pretrained language models (LMs) has predominantly focused on sentence-level syntactic tasks. In this paper, we introduce document-level discourse probing to evaluate the ability of pretrained LMs to capture document-level ...
Fajri Koto, Jey Han Lau, Tim Baldwin
semanticscholar +1 more source
Probing entanglement in a many-body–localized system [PDF]
Science, 2018 A logarithmic signature Some one-dimensional disordered interacting quantum systems have been theoretically predicted to display a property termed many-body localization (MBL), where the system retains the memory of its initial state and fails to ...
R. Schittko +8 more
semanticscholar +1 more source