Robust Natural Language Processing: Recent Advances, Challenges, and Future Directions
Open accessThe DOI matches its Crossref record; 28 of the 28 references it cites were found in a registry or an index.
Abstract
Recent natural language processing (NLP) techniques have accomplished high performance on benchmark data sets, primarily due to the significant improvement in the performance of deep learning. The advances in the research community have led to great enhancements in state-of-the-art production systems for NLP tasks, such as virtual assistants, speech recognition, and sentiment analysis. However, such NLP systems still often fail when tested with adversarial attacks. The initial lack of robustness exposed troubling gaps in current models’ language understanding capabilities, creating problems when NLP systems are deployed in real life. In this paper, we present a structured overview of NLP robustness research by summarizing the literature in a systemic way across various dimensions. We then take a deep-dive into the various dimensions of robustness, across techniques, metrics, embedding, and benchmarks. Finally, we argue that robustness should be multi-dimensional, provide insights into current research, identify gaps in the literature to suggest directions worth pursuing to address these gaps.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325 · cited 3×
“GLUE is designed for analyzing and effectively evaluating NLP systems on three linguistic tasks, classification (benchmark datasets: SST-2 and CoLA), paraphrasing (benchmark datasets: MRPC, STS-B, and QQP), and inference (benchmark datasets: MNLI, WNLI, QNLI, and RTE) [142].”
BackgroundExplaining and Harnessing Adversarial Examples
Goodfellow, I. J., Shlens, J., & Szegedy, C. (2014) · arXiv · doi:10.48550/arxiv.1412.6572 · Cited by 10,381
Towards Evaluating the Robustness of Neural Networks
Carlini, N., & Wagner, D. (2017) · 2017 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2017.49 · Cited by 8,859
Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank
Socher, R., Perelygin, A., Wu, J., Chuang, J., Manning, C. D., Ng, A., & Potts, C. (2013) · Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d13-1170 · Cited by 6,878
Towards Deep Learning Models Resistant to Adversarial Attacks
Madry, A., Makelov, A., Schmidt, L., Tsipras, D., & Vladu, A. (2017) · arXiv · doi:10.48550/arxiv.1706.06083 · Cited by 6,602
Know What You Don't Know: Unanswerable Questions for SQuAD
Rajpurkar, P., Jia, R., & Liang, P. (2018) · arXiv · doi:10.48550/arxiv.1806.03822 · Cited by 2,242
Theoretically principled trade-off between robustness and accuracy
Zhang, H., Yu, Y., Jiao, J., Xing, E. P., Ghaoui, L. E., & Jordan, M. I. (2019) · arXiv · doi:10.48550/arxiv.1901.08573 · Cited by 1,560
Lin, T., Wang, Y., Liu, X., & Qiu, X. (2022) · AI Open · doi:10.1016/j.aiopen.2022.10.001 · Cited by 1,422
Lin, T., Wang, Y., Liu, X., & Qiu, X. (2021) · arXiv · doi:10.48550/arxiv.2106.04554 · Cited by 1,420
Adversarial Examples for Evaluating Reading Comprehension Systems
Jia, R., & Liang, P. (2017) · Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d17-1215 · Cited by 1,410 · cited 6×
“This attack method has been used in numerous research studies [21], [70], [76], [77], [78], [79], [80].”
BackgroundMethodInfluentialPeng, Y., Yan, S., & Lu, Z. (2019) · Proceedings of the 18th BioNLP Workshop and Shared Task · doi:10.18653/v1/w19-5006 · Cited by 899
Beyond accuracy: behavioral testing of NLP models with CheckList
Ribeiro, M. T., Wu, T., Guestrin, C., & Singh, S. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.442 · Cited by 890
FastText.zip: Compressing text classification models
Joulin, A., Grave, E., Bojanowski, P., Douze, M., Jégou, H., & Mikolov, T. (2016) · arXiv · doi:10.48550/arxiv.1612.03651 · Cited by 887
Black-Box Generation of Adversarial Text Sequences to Evade Deep Learning Classifiers
Gao, J., Lanchantin, J., Soffa, M. L., & Qi, Y. (2018) · 2018 IEEE Security and Privacy Workshops (SPW) · doi:10.1109/spw.2018.00016 · Cited by 693
SummEval: Re-evaluating Summarization Evaluation
Fabbri, A. R., Kryściński, W., McCann, B., Xiong, C., Socher, R., & Radev, D. (2021) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00373 · Cited by 587 · cited 1×
“As a case in point, Amazon builtan NLP-based recruiting tool that was deemed as a failure (and was eventually scraped) because the NLP-powered model demonstrated bias against female applicants [30].”
BackgroundGenerating Natural Adversarial Examples
Zhao, Z., Dua, D., & Singh, S. (2017) · arXiv · doi:10.48550/arxiv.1710.11342 · Cited by 508 · cited 1×
“From an attack granularity perspective, various studies have been carried out using either a character-level, wordlevel, or sentence-level attack [37], [38], [45], [53], [58], [59], [75].”
MethodThe Natural Language Decathlon: Multitask Learning as Question Answering
McCann, B., Keskar, N. S., Xiong, C., & Socher, R. (2018) · arXiv · doi:10.48550/arxiv.1806.08730 · Cited by 448
Jang, B., Kim, M., Harerimana, G., Kang, S.-U., & Kim, J. W. (2020) · Applied Sciences · doi:10.3390/app10175841 · Cited by 442 · cited 1×
“The authors also use the generalization benchmark to train a model on the SST-2 [32] dataset and evaluate on the IMDB [33] (both of which are popular benchmark datasets for the sentiment analysis task).”
MethodBreaking NLI Systems with Sentences that Require Simple Lexical Inferences
Glockner, M., Shwartz, V., & Goldberg, Y. (2018) · Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) · doi:10.18653/v1/p18-2103 · Cited by 319 · cited 1×
“The classification accuracy has been utilized by numerous research works [34], [35], [40], [41], [45], [59], [103], [105], [106].”
MethodSocial Biases in NLP Models as Barriers for Persons with Disabilities
Hutchinson, B., Prabhakaran, V., Denton, E., Webster, K., Zhong, Y., & Denuyl, S. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.487 · Cited by 274
Social Biases in NLP Models as Barriers for Persons with Disabilities
Hutchinson, B., Prabhakaran, V., Denton, E., Webster, K., Zhong, Y., & Denuyl, S. (2020) · arXiv · doi:10.48550/arxiv.2005.00813 · Cited by 274 · cited 2×
“, variables or attributes) of a dataset are more heavily represented than others [144], this skewing the resulting NLP model recognizing such bias and affecting its operation against a more diverse data distribution.”
BackgroundFuzzy bag-of-words model for document representation
Zhao, R., & Mao, K. (2018) · IEEE Transactions on Fuzzy Systems · doi:10.1109/tfuzz.2017.2690222 · Cited by 252 · cited 1×
“The bag-of-words model, or BoW for short, is a technique used for extracting features from raw data for use in NLP models [132].”
MethodPredictive biases in natural language processing models: A conceptual framework and overview
Shah, D. S., Schwartz, H. A., & Hovy, D. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.468 · Cited by 241 · cited 1×
“The issue of dataset bias in NLP has been studied extensively in numerous research works [44], [46], [144], [145].”
BackgroundLearning General Purpose Distributed Sentence Representations via Large Scale Multi-task Learning
Subramanian, S., Trischler, A., Bengio, Y., & Pal, C. J. (2018) · arXiv · doi:10.48550/arxiv.1804.00079 · Cited by 207
Reducing Sentiment Bias in Language Models via Counterfactual Evaluation
Huang, P.-S., Zhang, H., Jiang, R., Stanforth, R., Welbl, J., Rae, J., Maini, V., Yogatama, D., & Kohli, P. (2020) · Findings of the Association for Computational Linguistics: EMNLP 2020 · doi:10.18653/v1/2020.findings-emnlp.7 · Cited by 124
Czarnowska, P., Vyas, Y., & Shah, K. (2021) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00425 · Cited by 72 · cited 1×
“The fairness metric has been utilized by various studies [108], [109], [113], [114], [121] to evaluate the degree of NLP robustness and ensure that large models are fair.”
MethodGrey-box Adversarial Attack And Defence For Sentiment Classification
Xu, Y., Zhong, X., Jimeno Yepes, A., & Lau, J. H. (2021) · Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies · doi:10.18653/v1/2021.naacl-main.321 · Cited by 54
Cited by
Gao, B., Wang, Y., Xie, H., Hu, Y., & Hu, Y. (2023) · Sage Open · doi:10.1177/21582440231210759 · Cited by 244
Large Language Models with Controllable Working Memory
Li, D., Rawat, A. S., Zaheer, M., Wang, X., Lukasik, M., Veit, A., Yu, F., & Kumar, S. (2023) · Findings of the Association for Computational Linguistics: ACL 2023 · doi:10.18653/v1/2023.findings-acl.112 · Cited by 46
Survey of Deep Learning Accelerators for Edge and Emerging Computing
Alam, S., Yakopcic, C., Wu, Q., Barnell, M., Khan, S., & Taha, T. M. (2024) · Electronics · doi:10.3390/electronics13152988 · Cited by 38
Detecting IoT Malware with Knowledge Distillation Technique
Jones, R., & Omar, M. (2023) · 2023 Congress in Computer Science, Computer Engineering, & Applied Computing (CSCE) · doi:10.1109/csce60160.2023.00026 · Cited by 13
Self-citation
Metadata
- Authors
- Omar, M., Choi, S., Nyang, D., & Mohaisen, D.
- Venue
- IEEE Access
- Published
- 2022
- Volume
- 10
- Pages
- 86038-86056
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W4293182797
- Cited by
- 85
- References
- 28
- FWCI
- 6.22
- Type
- journal article
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.