# Advancing the Robustness of Large Language Models through Self-Denoised Smoothing

arXiv: https://arxiv.org/abs/2404.12274 · HTML: https://mlsanity.com/paper/2404.12274/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Placed in Jailbreaking, one of 12 directions in LLM safety and security.
- Builds on 27 of its 36 references that are in the map.
- First posted on arXiv 18 Apr 2024.

## Placed in

- LLM safety and security / [Jailbreaking](https://mlsanity.com/safety/jailbreaking/)

## Builds on (most cited first, 27 of 27)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [Character-level Convolutional Networks for Text Classification](https://mlsanity.com/paper/1509.01626/) (2015)
- [Theoretically Principled Trade-off between Robustness and Accuracy](https://mlsanity.com/paper/1901.08573/) (2019)
- [Jailbreaking Black Box Large Language Models in Twenty Queries](https://mlsanity.com/paper/2310.08419/) (2023)
- [Certified Adversarial Robustness via Randomized Smoothing](https://mlsanity.com/paper/1902.02918/) (2019)
- [Galactica: A Large Language Model for Science](https://mlsanity.com/paper/2211.09085/) (2022)
- [Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailment](https://mlsanity.com/paper/1907.11932/) (2019)
- [SmoothLLM: Defending Large Language Models Against Jailbreaking Attacks](https://mlsanity.com/paper/2310.03684/) (2023)
- [TextBugger: Generating Adversarial Text Against Real-world Applications](https://mlsanity.com/paper/1812.05271/) (2018)
- [Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifiers](https://mlsanity.com/paper/1801.04354/) (2018)
- [Hard Prompts Made Easy: Gradient-Based Discrete Optimization for Prompt Tuning and Discovery](https://mlsanity.com/paper/2302.03668/) (2023)
- [FreeLB: Enhanced Adversarial Training for Natural Language Understanding](https://mlsanity.com/paper/1909.11764/) (2019)
- [Gradient-based Adversarial Attacks against Text Transformers](https://mlsanity.com/paper/2104.13733/) (2021)
- [Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models](https://mlsanity.com/paper/2111.02840/) (2021)
- [Contextualized Perturbation for Textual Adversarial Attack](https://mlsanity.com/paper/2009.07502/) (2020)
- [Empowering Molecule Discovery for Molecule-Caption Translation with Large Language Models: A ChatGPT Perspective](https://mlsanity.com/paper/2306.06615/) (2023)
- [Revisiting and Advancing Fast Adversarial Training Through The Lens of Bi-Level Optimization](https://mlsanity.com/paper/2112.12376/) (2021)
- [SAFER: A Structure-free Approach for Certified Robustness to Adversarial Word Substitutions](https://mlsanity.com/paper/2005.14424/) (2020)
- [Tight Certificates of Adversarial Robustness for Randomly Smoothed Classifiers](https://mlsanity.com/paper/1906.04948/) (2019)
- [Certified Robustness to Text Adversarial Attacks by Randomized [MASK]](https://mlsanity.com/paper/2105.03743/) (2021)
- [How to Robustify Black-Box ML Models? A Zeroth-Order Optimization Perspective](https://mlsanity.com/paper/2203.14195/) (2022)
- [Using Adversarial Attacks to Reveal the Statistical Bias in Machine Reading Comprehension Models](https://mlsanity.com/paper/2105.11136/) (2021)
- [Certified Robustness Against Natural Language Attacks by Causal Intervention](https://mlsanity.com/paper/2205.12331/) (2022)
- [Query-Efficient and Scalable Black-Box Adversarial Attacks on Discrete Sequential Data via Bayesian Optimization](https://mlsanity.com/paper/2206.08575/) (2022)
- [TextGrad: Advancing Robustness Evaluation in NLP by Gradient-Driven Optimization](https://mlsanity.com/paper/2212.09254/) (2022)

## Built on by (most cited first, 16 of 16)

- [Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond](https://mlsanity.com/paper/2502.05374/) (2025)
- [A safety realignment framework via subspace-oriented model fusion for large language models](https://mlsanity.com/paper/2405.09055/) (2024)
- [Overcoming the Retrieval Barrier: Indirect Prompt Injection in the Wild for LLM Systems](https://mlsanity.com/paper/2601.07072/) (2026)
- [Rethinking Jailbreak Detection of Large Vision Language Models with Representational Contrastive Scoring](https://mlsanity.com/paper/2512.12069/) (2025)
- [Enhancing Robustness of LLM-Driven Multi-Agent Systems through Randomized Smoothing](https://mlsanity.com/paper/2507.04105/) (2025)
- [CluCERT: Certifying LLM Robustness via Clustering-Guided Denoising Smoothing](https://mlsanity.com/paper/2512.08967/) (2025)
- [Enhancing the Capability and Robustness of Large Language Models through Reinforcement Learning-Driven Query Refinement](https://mlsanity.com/paper/2407.01461/) (2024)
- [Tougher Text, Smarter Models: Raising the Bar for Adversarial Defence Benchmarks](https://mlsanity.com/paper/2501.02654/) (2025)
- [Noise-Robustness Through Noise: A Framework combining Asymmetric LoRA with Poisoning MoE](https://mlsanity.com/paper/2505.23868/) (2025)
- [Adversarial Defence without Adversarial Defence: Enhancing Language Model Robustness via Instance-level Principal Component Removal](https://mlsanity.com/paper/2507.21750/) (2025)
- [Layered LLM Defenses as an Ensemble: Access Tiers, Inference Cost, and the Measured Failure Correlation Between Defense Layers](https://mlsanity.com/paper/2608.28327/) (2026)
- [DiffuseDef: Improved Robustness to Adversarial Attacks via Iterative Denoising](https://mlsanity.com/paper/2407.00248/) (2024)
- [Randomized Smoothing Meets Vision-Language Models](https://mlsanity.com/paper/2509.16088/) (2025)
- [Disentangling Intent from Role: Adversarial Self-Play for Persona-Invariant Safety Alignment](https://mlsanity.com/paper/2605.01899/) (2026)
- [AdaptDel: Adaptable Deletion Rate Randomized Smoothing for Certified Robustness](https://mlsanity.com/paper/2511.09316/) (2025)
- [Breaking the Assumptions: Auditing Input-Side Jailbreak Defenses Against Semantic Attacks](https://mlsanity.com/paper/2608.21895/) (2026)

## Abstract (arXiv)

Although large language models (LLMs) have achieved significant success, their vulnerability to adversarial perturbations, including recent jailbreak attacks, has raised considerable concerns. However, the increasing size of these models and their limited access make improving their robustness a challenging task. Among various defense strategies, randomized smoothing has shown great potential for LLMs, as it does not require full access to the model's parameters or fine-tuning via adversarial training. However, randomized smoothing involves adding noise to the input before model prediction, and the final model's robustness largely depends on the model's performance on these noise corrupted data. Its effectiveness is often limited by the model's sub-optimal performance on noisy data. To address this issue, we propose to leverage the multitasking nature of LLMs to first denoise the noisy inputs and then to make predictions based on these denoised versions. We call this procedure self-denoised smoothing. Unlike previous denoised smoothing techniques in computer vision, which require training a separate model to enhance the robustness of LLMs, our method offers significantly better efficiency and flexibility. Our experimental results indicate that our method surpasses existing methods in both empirical and certified robustness in defending against adversarial attacks for both downstream tasks and human alignments (i.e., jailbreak attacks). Our code is publicly available at https://github.com/UCSB-NLP-Chang/SelfDenoise