No abstract
The robustness of Text-to-SQL parsers against adversarial perturbations plays a crucial role in delivering highly reliable applications. Previous studies along this line primarily focused on perturbations in the natural language question side, neglecting the variability of tables. Motivated by this, we propose the Adversarial Table Perturbation (ATP) as a new attacking paradigm to measure the robustness of Textto-SQL models. Following this proposition, we curate ADVETA, the first robustness evaluation benchmark featuring natural and realistic ATPs. All tested state-of-the-art models experience dramatic performance drops on ADVETA, revealing models' vulnerability in real-world practices. To defend against ATP, we build a systematic adversarial training example generation framework tailored for better contextualization of tabular data. Experiments show that our approach not only brings the best robustness improvement against tableside perturbations but also substantially empowers models against NL-side perturbations. We release our benchmark and code at: https://github.com/microsoft/ContextualSP.
Background: With the rapid development of power system, oil-immersed transformers are widely used in the substation and distribution system. The faults of oil-immersed transformers are large threat to the power system. Therefore, it is significant that the faults of oil-immersed transformers can be diagnosed accurately. Objective: To accurately diagnose the faults of oil-immersed transformers through machine learning methods and swarm intelligent algorithms. Methods: To accurately diagnose the faults of oil-immersed transformers, a fault diagnosis method based on T-distributed stochastic neighbor embedding and support vector machine is proposed. The improved beetle antennae search algorithm is used to optimize the parameters of support vector machine. Firstly, the non-coding ratio method is used to obtain nine-dimensional characteristic indices. Secondly, the original nine-dimensional data are reduced to three-dimensional by T-distributed stochastic neighbor embedding. Lastly, the data after dimensionality reduction are used as the input of the support vector machine optimized by improved beetle antennae search algorithm and the fault types of transformers can be diagnosed. Results: The accuracy rate is 94.53% and the operation time is about 1.88s. The results indicate that the method proposed by this paper is reasonable. Conclusion: The experimental results show that the method proposed by this paper has a high accuracy rate and low operation time. Mixed faults that are difficult to diagnose also can be diagnosed by this paper's method. In the era of big data, there is a lot of data of transformer, so the method proposed in this paper has certain engineering significance.
We present LogiGAN, an unsupervised adversarial pre-training framework for improving logical reasoning abilities of language models. Upon automatic identifying logical reasoning phenomena in massive text corpus via detection heuristics, we train language models to predict the masked-out logical statements. Inspired by the facilitation effect of reflective thinking in human learning, we analogically simulate the learning-thinking process with an adversarial Generator-Verifier architecture to assist logic learning. LogiGAN implements a novel sequential GAN approach that (a) circumvents the non-differentiable challenge of the sequential GAN by leveraging the Generator as a sentence-level generative likelihood scorer with a learning objective of reaching scoring consensus with the Verifier; (b) is computationally feasible for large-scale pre-training with arbitrary target length. Both base and large size language models pre-trained with LogiGAN demonstrate obvious performance improvement on 12 datasets requiring general reasoning abilities, revealing the fundamental role of logic in broad reasoning, as well as the effectiveness of LogiGAN. Ablation studies on LogiGAN components reveal the relative orthogonality between linguistic and logic abilities and suggest that reflective thinking's facilitation effect might also generalize to machine learning.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.