mlsanityData to 14 Oct 2026

Randomized Smoothing Meets Vision-Language Models

arXiv 2509.16088 ↗First posted 19 Sep 2025cs.LGcited by 1

On the map: Multimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness

Abstract (arXiv)

Randomized smoothing (RS) is one of the prominent techniques to ensure the correctness of machine learning models, where point-wise robustness certificates can be derived analytically. While RS is well understood for classification, its application to generative models is unclear, since their outputs are sequences rather than labels. We resolve this by connecting generative outputs to an oracle classification task and showing that RS can still be enabled: the final response can be classified as a discrete action (e.g., service-robot commands in VLAs), as harmful vs. harmless (content moderation or toxicity detection in VLMs), or even applying oracles to cluster answers into semantically equivalent ones. Provided that the error rate for the oracle classifier comparison is bounded, we develop the theory that associates the number of samples with the corresponding robustness radius. We further derive improved scaling laws analytically relating the certified radius and accuracy to the number of samples, showing that the earlier result of 2 to 3 orders of magnitude fewer samples sufficing with minimal loss remains valid even under weaker assumptions. Together, these advances make robustness certification both well-defined and computationally feasible for state-of-the-art VLMs, as validated against recent jailbreak-style adversarial attacks.

Key facts. Cited by 1 paper 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 27 of its 40 references that are in the map. First posted on arXiv 19 Sep 2025.

Builds on

27 of 40 references.

References · 27

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2023GPT-4 Technical Reportcited by 18,396
  4. 2024The Llama 3 Herd of Modelscited by 14,715
  5. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  6. 2013Intriguing properties of neural networkscited by 5,155
  7. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
  8. 2021On the Opportunities and Risks of Foundation Modelscited by 4,033
  9. 2024Gemini 1.5: Unlocking multimodal understanding across millions of tokens of contextcited by 3,224
  10. 2024Qwen2 Technical Reportcited by 2,120
  11. 2024Gemma 2: Improving Open Language Models at a Practical Sizecited by 1,998
  12. 2022Fast Inference from Transformers via Speculative Decodingcited by 1,653
15 more
  1. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,340
  2. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,268
  3. 2017Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networkscited by 744
  4. 2013Speech Recognition with Deep Recurrent Neural Networkscited by 669
  5. 2017Provable defenses against adversarial examples via the convex outer adversarial polytopecited by 567
  6. 2019Provably Robust Deep Learning via Adversarially Trained Smoothed Classifierscited by 330
  7. 2018On the Effectiveness of Interval Bound Propagation for Training Verifiably Robust Modelscited by 328
  8. 2024An Introduction to Vision-Language Modelingcited by 178
  9. 2020Randomized Smoothing of All Shapes and Sizescited by 147
  10. 2020Certified Defense to Image Transformations via Randomized Smoothingcited by 51
  11. 2021Scalable Certified Segmentation via Randomized Smoothingcited by 35
  12. 2024Advancing the Robustness of Large Language Models through Self-Denoised Smoothingcited by 16
  13. 2021Input-Specific Robustness Certification for Randomized Smoothingcited by 11
  14. 2024Treatment of Statistical Estimation Problems in Randomized Smoothing for Adversarial Robustnesscited by 8
  15. 2024Estimating the Robustness Radius for Randomized Smoothing with 100× Sample Efficiencycited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Certified Robustness under Heterogeneous Perturbations via Hybrid Randomized Smoothingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2021Improved, Deterministic Smoothing for L_1 Certified Robustness
  2. 2023Incremental Randomized Smoothing Certification
  3. 2025Robust Representation Consistency Model via Contrastive Denoising
  4. 2023Machine Learning needs Better Randomness Standards: Randomised Smoothing and PRNG-based attacks
  5. 2025Robustness of Probabilistic Models to Low-Quality Data: A Multi-Perspective Analysis
  6. 2024Your Diffusion Model is Secretly a Certifiably Robust Classifier
  7. 2024PromptSmooth: Certifying Robustness of Medical Vision-Language Models via Prompt Learning
  8. 2025Smoothed Embeddings for Robust Language Models