Related papers: Avoiding the Hypothesis-Only Bias in Natural Langu…

Natural Language Inference (NLI) is the task of determining whether a premise entails, contradicts, or is neutral with respect to a given hypothesis. The task is often framed as emulating human inferential processes, in which commonsense…

Over the last years, various sentence embedders have been an integral part in the success of current machine learning approaches to Natural Language Processing (NLP). Unfortunately, multiple sources have shown that the bias, inherent in the…

Fundamental questions remain about when and why adversarial examples arise in neural networks, with competing views characterising them either as artifacts of the irregularities in the decision landscape or as products of sensitivity to…

Gender-bias stereotypes have recently raised significant ethical concerns in natural language processing. However, progress in detection and evaluation of gender bias in natural language understanding through inference is limited and…

Traditional (fickle) adversarial examples involve finding a small perturbation that does not change an input's true label but confuses the classifier into outputting a different prediction. Conversely, obstinate adversarial examples occur…

Natural language inference (NLI) is formulated as a unified framework for solving various NLP problems such as relation extraction, question answering, summarization, etc. It has been studied intensively in the past few years thanks to the…

Large neural models have demonstrated human-level performance on language and vision benchmarks, while their performance degrades considerably on adversarial or out-of-distribution samples. This raises the question of whether these models…

It has been demonstrated that deep neural networks are prone to noisy examples particular adversarial samples during inference process. The gap between robust deep learning systems in real world applications and vulnerable neural networks…

Adversarial attacks have been shown to be highly effective at degrading the performance of deep neural networks (DNNs). The most prominent defense is adversarial training, a method for learning a robust model. Nevertheless, adversarial…

With the development of high computational devices, deep neural networks (DNNs), in recent years, have gained significant popularity in many Artificial Intelligence (AI) applications. However, previous efforts have shown that DNNs were…

Despite pre-trained language models have proven useful for learning high-quality semantic representations, these models are still vulnerable to simple perturbations. Recent works aimed to improve the robustness of pre-trained models mainly…

Audio events are quite often overlapping in nature, and more prone to noise than visual signals. There has been increasing evidence for the superior performance of representations learned using sparse dictionaries for applications like…

In cross-lingual text classification, one seeks to exploit labeled data from one language to train a text classification model that can then be applied to a completely different language. Recent multilingual representation models have made…

Deep neural networks are vulnerable to adversarial examples, i.e., carefully-perturbed inputs aimed to mislead classification. This work proposes a detection method based on combining non-linear dimensionality reduction and density…

Natural language inference (NLI) is the task of determining if a natural language hypothesis can be inferred from a given premise in a justifiable manner. NLI was proposed as a benchmark task for natural language understanding. Existing…

In adversarial imitation learning, a discriminator is trained to differentiate agent episodes from expert demonstrations representing the desired behavior. However, as the trained policy learns to be more successful, the negative examples…

While deep learning in the form of recurrent neural networks (RNNs) has caused a significant improvement in neural language modeling, the fact that they are extremely prone to overfitting is still a mainly unresolved issue. In this paper we…

Many datasets have been shown to contain incidental correlations created by idiosyncrasies in the data collection process. For example, sentence entailment datasets can have spurious word-class correlations if nearly all contradiction…

Presence of bias (in datasets or tasks) is inarguably one of the most critical challenges in machine learning applications that has alluded to pivotal debates in recent years. Such challenges range from spurious associations between…

Success in natural language inference (NLI) should require a model to understand both lexical and compositional semantics. However, through adversarial evaluation, we find that several state-of-the-art models with diverse architectures are…