# SmoothGuard: Defending Multimodal Large Language Models with Noise Perturbation and Clustering Aggregation

arXiv: https://arxiv.org/abs/2510.26830 · HTML: https://mlsanity.com/paper/2510.26830/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 29 of its 31 references that are in the map.
- First posted on arXiv 29 Oct 2025.

## Placed in

- Multimodal LLMs / vision-language models / [MLLM jailbreaks and adversarial robustness](https://mlsanity.com/mllm/mllm-jailbreaks-and-adversarial-robustness/)

## Builds on (most cited first, 29 of 29)

- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Intriguing properties of neural networks](https://mlsanity.com/paper/1312.6199/) (2013)
- [Certified Adversarial Robustness via Randomized Smoothing](https://mlsanity.com/paper/1902.02918/) (2019)
- [Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations](https://mlsanity.com/paper/2312.06674/) (2023)
- [Ensemble Adversarial Training: Attacks and Defenses](https://mlsanity.com/paper/1705.07204/) (2017)
- [mPLUG-Owl2: Revolutionizing Multi-modal Large Language Model with Modality Collaboration](https://mlsanity.com/paper/2311.04257/) (2023)
- [MLLM-as-a-Judge: Assessing Multimodal LLM-as-a-Judge with Vision-Language Benchmark](https://mlsanity.com/paper/2402.04788/) (2024)
- [Visual Adversarial Examples Jailbreak Aligned Large Language Models](https://mlsanity.com/paper/2306.13213/) (2023)
- [On Evaluating Adversarial Robustness of Large Vision-Language Models](https://mlsanity.com/paper/2305.16934/) (2023)
- [MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Models](https://mlsanity.com/paper/2311.17600/) (2023)
- [MLLM-Protector: Ensuring MLLM's Safety without Hurting Performance](https://mlsanity.com/paper/2401.02906/) (2024)
- [Safety of Multimodal Large Language Models on Images and Texts](https://mlsanity.com/paper/2402.00357/) (2024)
- [A Survey of Robust Adversarial Training in Pattern Recognition: Fundamental, Theory, and Methodologies](https://mlsanity.com/paper/2203.14046/) (2022)
- [A Unified Gradient Regularization Family for Adversarial Examples](https://mlsanity.com/paper/1511.06385/) (2015)
- [MIBench: Evaluating Multimodal Large Language Models over Multiple Images](https://mlsanity.com/paper/2407.15272/) (2024)
- [Cantor: Inspiring Multimodal Chain-of-Thought of MLLM](https://mlsanity.com/paper/2404.16033/) (2024)
- [MathAttack: Attacking Large Language Models Towards Math Solving Ability](https://mlsanity.com/paper/2309.01686/) (2023)
- [Safeguarding Vision-Language Models Against Patched Visual Prompt Injectors](https://mlsanity.com/paper/2405.10529/) (2024)
- [From LLMs to MLLMs: Exploring the Landscape of Multimodal Jailbreaking](https://mlsanity.com/paper/2406.14859/) (2024)
- [Audio Jailbreak: An Open Comprehensive Benchmark for Jailbreaking Large Audio-Language Models](https://mlsanity.com/paper/2505.15406/) (2025)
- [SpeechGuard: Exploring the Adversarial Robustness of Multimodal Large Language Models](https://mlsanity.com/paper/2405.08317/) (2024)
- [VLMGuard: Bootstrapping Malicious Prompt Detectors from Unlabeled Vision-Language Prompts in the Wild](https://mlsanity.com/paper/2410.00296/) (2024)
- [AudioJailbreak: Jailbreak Attacks against End-to-End Large Audio-Language Models](https://mlsanity.com/paper/2505.14103/) (2025)
- [SPIRIT: Patching Speech Language Models against Jailbreak Attacks](https://mlsanity.com/paper/2505.13541/) (2025)
- [BaThe: Defense against the Jailbreak Attack in Multimodal Large Language Models by Treating Harmful Instruction as Backdoor Trigger](https://mlsanity.com/paper/2408.09093/) (2024)
- [Audio Jailbreak Attacks: Exposing Vulnerabilities in SpeechGPT in a White-Box Framework](https://mlsanity.com/paper/2505.18864/) (2025)
- [Investigating Vulnerabilities and Defenses Against Audio-Visual Attacks: A Comprehensive Survey Emphasizing Multimodal Models](https://mlsanity.com/paper/2506.11521/) (2025)

## Built on by (most cited first, 2 of 2)

- [Audio Jailbreaks in Large Audio-Language Models: Taxonomy, Attack-Defense Analysis, and Cost-Aware Evaluation](https://mlsanity.com/paper/2605.30031/) (2026)
- [Clustered Randomized Smoothing for Stochastic Prediction Functions](https://mlsanity.com/paper/2608.12037/) (2026)

## Abstract (arXiv)

Multimodal large language models (MLLMs) have achieved impressive performance across diverse tasks by jointly reasoning over textual and visual inputs. Despite their success, these models remain highly vulnerable to adversarial manipulations, raising concerns about their safety and reliability in deployment. In this work, we first generalize an approach for generating adversarial images within the HuggingFace ecosystem and then introduce SmoothGuard, a lightweight and model-agnostic defense framework that enhances the robustness of MLLMs through randomized noise injection and clustering-based prediction aggregation. Our method perturbs continuous modalities (e.g., images and audio) with Gaussian noise, generates multiple candidate outputs, and applies embedding-based clustering to filter out adversarially influenced predictions. The final answer is selected from the majority cluster, ensuring stable responses even under malicious perturbations. Extensive experiments on POPE, LLaVA-Bench (In-the-Wild), and MM-SafetyBench demonstrate that SmoothGuard improves resilience to adversarial attacks while maintaining competitive utility. Ablation studies further identify an optimal noise range (0.1-0.2) that balances robustness and utility.