paper-with-me

홈 › Papers

On Adversarial Examples for Biomedical NLP Tasks

2020-04-23 · Vladimir Araujo, Andres Carvallo, Carlos Aspillaga, Denis Parra

The success of pre-trained word embeddings has motivated its use in tasks in the biomedical domain. The BERT language model has shown remarkable results on standard performance metrics in tasks such as Named Entity Recognition (NER) and Semantic Textual Similarity (STS), which has brought significant progress in the field of NLP. However, it is unclear whether these systems work seemingly well in critical domains, such as legal or medical. For that reason, in this work, we propose an adversarial evaluation scheme on two well-known datasets for medical NER and STS. We propose two types of attacks inspired by natural spelling errors and typos made by humans. We also propose another type of attack that uses synonyms of medical terms. Under these adversarial settings, the accuracy of the models drops significantly, and we quantify the extent of this performance loss. We also show that we can significantly improve the robustness of the models by training them with adversarial examples. We hope our work will motivate the use of adversarial examples to evaluate and develop models with increased robustness for medical tasks.

📄 PDF Abstract BibTeX arXiv:2004.11157

Code (0)

등록된 구현이 없습니다.

Tasks

Language ModelingLanguage Modellingnamed-entity-recognitionNamed Entity RecognitionNamed Entity Recognition (NER)NERSemantic Textual SimilaritySTSWord Embeddings

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Residual Connection 설명 없음
Attention Dropout Attention Dropout is a type of dropout used in attention-based architectures, where elements are randomly dropped out of the…
Linear Warmup With Linear Decay Linear Warmup With Linear Decay is a learning rate schedule in which we increase the learning rate linearly for $n$ updates and then linearly decay afterwards.
Weight Decay 설명 없음
Refunds@Expedia|||How do I get a full refund from Expedia? “How do I get a full refund from Expedia? How do I get a full refund from Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Quick Help &…
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Adam 설명 없음

Similar Papers 제목 키워드 기반

Adversarial Evaluation of BERT for Biomedical Named Entity Recognition

2020-07-01 · WS 2020 7 · Vladimir Araujo, Andr{\'e}s Carvallo, Denis Parra

The success of pre-trained word embeddings of the BERT model has motivated its use in tasks in the biomedical domain. However, it is not clear if this model works correctly in real scenarios. In this work, we propose an …

named-entity-recognitionNamed Entity RecognitionNamed Entity Recognition (NER)Word Embeddings

Detecting Adversarial Examples in Batches -- a geometrical approach

2022-06-17 · Danush Kumar Venkatesh, Peter Steinbach

Many deep learning methods have successfully solved complex tasks in computer vision and speech recognition applications. Nonetheless, the robustness of these models has been found to be vulnerable to perturbed inputs or…

Adversarial AttackAdversarial Attack DetectionImage Classification

Stress Test Evaluation of Biomedical Word Embeddings

2021-07-24 · NAACL (BioNLP) 2021 6 · Vladimir Araujo, Andrés Carvallo, Carlos Aspillaga, Camilo Thorne 외

The success of pretrained word embeddings has motivated their use in the biomedical domain, with contextualized embeddings yielding remarkable results in several biomedical NLP tasks. However, there is a lack of research…

named-entity-recognitionNamed Entity RecognitionNamed Entity Recognition (NER)NER+1

Impact of Adversarial Examples on Deep Learning Models for Biomedical Image Segmentation

2019-07-30 · Utku Ozbulak, Arnout Van Messem, Wesley De Neve

Deep learning models, which are increasingly being used in the field of medical image analysis, come with a major security risk, namely, their vulnerability to adversarial examples. Adversarial examples are carefully cra…

Image SegmentationLesion SegmentationMedical Image AnalysisOptic Disc Segmentation+3

Assessing biomedical knowledge robustness in large language models by query-efficient sampling attacks

2024-02-16 · R. Patrick Xian, Alex J. Lee, Satvik Lolla, Vincent Wang 외

The increasing depth of parametric domain knowledge in large language models (LLMs) is fueling their rapid deployment in real-world applications. Understanding model vulnerabilities in high-stakes and knowledge-intensive…

Distractor GenerationQuestion Answering