mlsanityData to 7 Oct 2026

Adversarial Training for Large Neural Language Models

arXiv 2004.08994 ↗First posted 20 Apr 2020cs.CLcited by 109

Not on the map. Why?

Abstract (arXiv)

Generalization and robustness are both key desiderata for designing machine learning methods. Adversarial training can enhance robustness, but past work often finds it hurts generalization. In natural language processing (NLP), pre-training large neural language models such as BERT have demonstrated impressive gain in generalization for a variety of tasks, with further improvement from adversarial fine-tuning. However, these models are still vulnerable to adversarial attacks. In this paper, we show that adversarial pre-training can improve both generalization and robustness. We propose a general algorithm ALUM (Adversarial training for large neural LangUage Models), which regularizes the training objective by applying perturbations in the embedding space that maximizes the adversarial loss. We present the first comprehensive study of adversarial training in all stages, including pre-training from scratch, continual pre-training on a well-trained model, and task-specific fine-tuning. ALUM obtains substantial gains over BERT on a wide range of NLP tasks, in both regular and adversarial scenarios. Even for models that have been well trained on extremely large text corpora, such as RoBERTa, ALUM can still produce significant gains from continual pre-training, whereas conventional non-adversarial methods can not. ALUM can be further combined with task-specific fine-tuning to attain additional gains. The ALUM code is publicly available at https://github.com/namisan/mt-dnn.

Key facts. Cited by 109 papers in the map (data through 7 Oct 2026). First posted on arXiv 20 Apr 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 27; 2022: 25; 2023: 25; 2024: 22; 2025: 8; 2026: 2.

109 papers.

Sort:

By area:

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2024HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusalcited by 1,330
  3. 2022LaMDA: Language Models for Dialog Applicationscited by 1,218
  4. 2023A Comprehensive Overview of Large Language Modelscited by 639
  5. 2023Prompt Injection attack against LLM-integrated Applicationscited by 568
  6. 2023GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Promptscited by 520
  7. 2023SmoothLLM: Defending Large Language Models Against Jailbreaking Attackscited by 434
  8. 2023A Survey on Large Language Model (LLM) Security and Privacy: The Good, the Bad, and the Uglycited by 380
  9. 2022Making the Most of Text Semantics to Improve Biomedical Vision--Language Processingcited by 336
  10. 2024Personal LLM Agents: Insights and Survey about the Capability, Efficiency and Securitycited by 257
  11. 2021Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Modelscited by 227
  12. 2021Muppet: Massive Multi-task Representations with Pre-Finetuningcited by 192
38 more
  1. 2024Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMscited by 158
  2. 2024Black-Box Access is Insufficient for Rigorous AI Auditscited by 123
  3. 2024Large Language Model Unlearning via Embedding-Corrupted Promptscited by 112
  4. 2021A Survey on Data Augmentation for Text Classificationcited by 111
  5. 2021Measure and Improve Robustness in NLP Models: A Surveycited by 107
  6. 2024Efficient Adversarial Training in LLMs with Continuous Attackscited by 104
  7. 2024Soft Prompt Threats: Attacking Safety Alignment and Unlearning in Open-Source LLMs through the Embedding Spacecited by 88
  8. 2024Defending Against Unforeseen Failure Modes with Latent Adversarial Trainingcited by 82
  9. 2021Towards Improving Adversarial Training of NLP Modelscited by 79
  10. 2024Building Guardrails for Large Language Modelscited by 61
  11. 2021Consistency Regularization for Cross-Lingual Fine-Tuningcited by 41
  12. 2023Hijacking Large Language Models via Adversarial In-Context Learningcited by 41
  13. 2021The Dangers of Underclaiming: Reasons for Caution When Reporting How NLP Systems Failcited by 38
  14. 2023SentimentGPT: Exploiting GPT for Advanced Sentiment Analysis and its Departure from Current Machine Learningcited by 38
  15. 2023Prompt Optimization via Adversarial In-Context Learningcited by 33
  16. 2023Token-Level Adversarial Prompt Detection Based on Perplexity Measures and Contextual Informationcited by 32
  17. 2021HiddenCut: Simple Data Augmentation for Natural Language Understanding with Better Generalizationcited by 31
  18. 2024On the Challenges and Opportunities in Generative AIcited by 31
  19. 2024PICLe: Eliciting Diverse Behaviors from Large Language Models with Persona In-Context Learningcited by 28
  20. 2022A Survey of Adversarial Defences and Robustness in NLPcited by 22
  21. 2022METRO: Efficient Denoising Pretraining of Large Scale Autoencoding Language Models with Model Generated Signalscited by 20
  22. 2023METAL: Metamorphic Testing Framework for Analyzing Large-Language Model Qualitiescited by 20
  23. 2022Improving Contrastive Learning of Sentence Embeddings with Case-Augmented Positives and Retrieved Negativescited by 16
  24. 2022Debiasing Masks: A New Framework for Shortcut Mitigation in NLUcited by 16
  25. 2021Adversarial Reinforced Instruction Attacker for Robust Vision-Language Navigationcited by 15
  26. 2021Distributed Methods with Compressed Communication for Solving Variational Inequalities, with Theoretical Guaranteescited by 15
  27. 2022On Robust Prefix-Tuning for Text Classificationcited by 15
  28. 2024Estimating the Probabilities of Rare Outputs in Language Modelscited by 15
  29. 2024Detecting and Understanding Vulnerabilities in Language Models via Mechanistic Interpretabilitycited by 13
  30. 2023A Review of ChatGPT Applications in Education, Marketing, Software Engineering, and Healthcare: Benefits, Drawbacks, and Research Directionscited by 12
  31. 2023VIBE: Topic-Driven Temporal Adaptation for Twitter Classificationcited by 12
  32. 2024A Comprehensive Survey of Contamination Detection Methods in Large Language Modelscited by 12
  33. 2021Unifying Model Explainability and Robustness for Joint Text Classification and Rationale Extractioncited by 11
  34. 2023Toward Adversarial Training on Contextualized Language Representationcited by 10
  35. 2023Improving the Robustness of Transformer-based Large Language Models with Dynamic Attentioncited by 10
  36. 2021Consistency Training with Virtual Adversarial Discrete Perturbationcited by 9
  37. 2021Understanding and Improvement of Adversarial Training for Network Embedding from an Optimization Perspectivecited by 9
  38. 2022Impact of Adversarial Training on Robustness and Generalizability of Language Modelscited by 9

50 most cited of 109 All 109 (JSON)