 Gradient-based Adversarial Attacks against Text TransformersConference on Empirical Methods in Natural Language Processing (EMNLP), 2021 |
 Consistency Training with Virtual Adversarial Discrete PerturbationNorth American Chapter of the Association for Computational Linguistics (NAACL), 2021 |
 Detoxifying Language Models Risks Marginalizing Minority VoicesNorth American Chapter of the Association for Computational Linguistics (NAACL), 2021 |
 Evaluating Pre-Trained Models for User Feedback Analysis in Software
Engineering: A Study on Classification of App-ReviewsEmpirical Software Engineering (EMSE), 2021 |
![Factual Probing Is [MASK]: Learning vs. Learning to Recall](https://cdn.parameterlab.de/resized-images/aHR0cHM6Ly9jZG4ucGFyYW1ldGVybGFiLmRlL3BhcGVycy8yMTA0LjA1MjQwL3BhZ2VzL3BhZ2UtMS1xODBkMzAwLmpwZw?w=256&q=75&f=webp) Factual Probing Is [MASK]: Learning vs. Learning to RecallNorth American Chapter of the Association for Computational Linguistics (NAACL), 2021 |
 Double Perturbation: On the Robustness of Robustness and Counterfactual
Bias EvaluationNorth American Chapter of the Association for Computational Linguistics (NAACL), 2021 |
 FUDGE: Controlled Text Generation With Future DiscriminatorsNorth American Chapter of the Association for Computational Linguistics (NAACL), 2021 |
 Achieving Model Robustness through Discrete Adversarial TrainingConference on Empirical Methods in Natural Language Processing (EMNLP), 2021 |
 Connecting Attributions and QA Model Behavior on Realistic
CounterfactualsConference on Empirical Methods in Natural Language Processing (EMNLP), 2021 |
 Dynabench: Rethinking Benchmarking in NLPNorth American Chapter of the Association for Computational Linguistics (NAACL), 2021 |
 |
 Plug-and-Blend: A Framework for Controllable Story Generation with
Blended Control CodesArtificial Intelligence and Interactive Digital Entertainment Conference (AIIDE), 2021 |
 Code-Mixing on Sesame Street: Dawn of the Adversarial PolyglotsNorth American Chapter of the Association for Computational Linguistics (NAACL), 2021 |
 Get Your Vitamin C! Robust Fact Verification with Contrastive EvidenceNorth American Chapter of the Association for Computational Linguistics (NAACL), 2021 |
 MERMAID: Metaphor Generation with Symbolism and Discriminative DecodingNorth American Chapter of the Association for Computational Linguistics (NAACL), 2021 |
 ENTRUST: Argument Reframing with Language Models and EntailmentNorth American Chapter of the Association for Computational Linguistics (NAACL), 2021 |
 T-Miner: A Generative Approach to Defend Against Trojan Attacks on
DNN-based Text ClassificationUSENIX Security Symposium (USENIX Security), 2021 |
 A Survey On Universal Adversarial AttackInternational Joint Conference on Artificial Intelligence (IJCAI), 2021 |
 Certified Robustness to Programmable Transformations in LSTMsConference on Empirical Methods in Natural Language Processing (EMNLP), 2021 |
 Model Agnostic Answer Reranking System for Adversarial Question
AnsweringConference of the European Chapter of the Association for Computational Linguistics (EACL), 2021 |
 BOLD: Dataset and Metrics for Measuring Biases in Open-Ended Language
GenerationConference on Fairness, Accountability and Transparency (FAccT), 2021 |
 Adversarial Stylometry in the Wild: Transferable Lexical Substitution
Attacks on Author ProfilingConference of the European Chapter of the Association for Computational Linguistics (EACL), 2021 |
 Generating Syntactically Controlled Paraphrases without Using Annotated
Parallel PairsConference of the European Chapter of the Association for Computational Linguistics (EACL), 2021 |
 Adv-OLM: Generating Textual Adversaries via OLMConference of the European Chapter of the Association for Computational Linguistics (EACL), 2021 |
 Data-to-text Generation by Splicing Together Nearest NeighborsConference on Empirical Methods in Natural Language Processing (EMNLP), 2021 |
 Persistent Anti-Muslim Bias in Large Language ModelsAAAI/ACM Conference on AI, Ethics, and Society (AIES), 2021 |
 BERT & Family Eat Word Salad: Experiments with Text UnderstandingAAAI Conference on Artificial Intelligence (AAAI), 2021 |
 DynaSent: A Dynamic Benchmark for Sentiment AnalysisAnnual Meeting of the Association for Computational Linguistics (ACL), 2020 |
 Generating Natural Language Attacks in a Hard Label Black Box SettingAAAI Conference on Artificial Intelligence (AAAI), 2020 |
 Analysis of Dominant Classes in Universal Adversarial PerturbationsKnowledge-Based Systems (KBS), 2020 |
 To what extent do human explanations of model behavior align with actual
model behavior?BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP (BlackboxNLP), 2020 |
 A Distributional Approach to Controlled Text GenerationInternational Conference on Learning Representations (ICLR), 2020 |
 AdvExpander: Generating Natural Language Adversarial Examples by
Expanding TextIEEE/ACM Transactions on Audio Speech and Language Processing (TASLP), 2020 |
 Multilingual Transfer Learning for QA Using Translation as Data
AugmentationAAAI Conference on Artificial Intelligence (AAAI), 2020 |
 On the Transferability of Adversarial Attacksagainst Neural Text
ClassifierConference on Empirical Methods in Natural Language Processing (EMNLP), 2020 |
 Automatic Detection of Machine Generated Text: A Critical SurveyInternational Conference on Computational Linguistics (COLING), 2020 |
 Leveraging Extracted Model Adversaries for Improved Black Box AttacksBlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP (BlackboxNLP), 2020 |
 AutoPrompt: Eliciting Knowledge from Language Models with Automatically
Generated PromptsConference on Empirical Methods in Natural Language Processing (EMNLP), 2020 |
