mlsanityData to 14 Oct 2026

Partially Recentralization Softmax Loss for Vision-Language Models Robustness

arXiv 2402.03627 ↗First posted 6 Feb 2024cs.CL cs.AIcited by 0

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

Abstract (arXiv)

As Large Language Models make a breakthrough in natural language processing tasks (NLP), multimodal technique becomes extremely popular. However, it has been shown that multimodal NLP are vulnerable to adversarial attacks, where the outputs of a model can be dramatically changed by a perturbation to the input. While several defense techniques have been proposed both in computer vision and NLP models, the multimodal robustness of models have not been fully explored. In this paper, we study the adversarial robustness provided by modifying loss function of pre-trained multimodal models, by restricting top K softmax outputs. Based on the evaluation and scoring, our experiments show that after a fine-tuning, adversarial robustness of pre-trained models can be significantly improved, against popular attacks. Further research should be studying, such as output diversity, generalization and the robustness-performance trade-off of this kind of loss functions. Our code will be available after this paper is accepted

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 32 of its 39 references that are in the map. First posted on arXiv 6 Feb 2024.

Builds on

32 of 39 references.

Closest ancestors · 2

Same direction, newest first

  1. 2023SA-Attack: Improving Adversarial Transferability of Vision-Language Pre-training Models via Self-Augmentation8 Dec
  2. 2022Towards Adversarial Attack on Vision-Language Pre-training Models19 Jun

Other references · 30

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  4. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  5. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  6. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  7. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  8. 2013Intriguing properties of neural networkscited by 5,155
  9. 2022BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generationcited by 4,322
  10. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,512
  11. 2016Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotationscited by 2,656
  12. 2014CIDEr: Consensus-based Image Description Evaluationcited by 2,046
18 more
  1. 2016Adversarial examples in the physical worldcited by 1,769
  2. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,517
  3. 2018Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labelscited by 997
  4. 2018Robustness May Be at Odds with Accuracycited by 880
  5. 2015Distillation as a Defense to Adversarial Perturbations against Deep Neural Networkscited by 678
  6. 2019Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailmentcited by 671
  7. 2020Overfitting in adversarially robust deep learningcited by 506
  8. 2020BERT-ATTACK: Adversarial Attack Against BERT Using BERTcited by 491
  9. 2018Generating Natural Language Adversarial Examplescited by 424
  10. 2020Bag of Tricks for Adversarial Trainingcited by 200
  11. 2014Towards Deep Neural Network Architectures Robust to Adversarial Examplescited by 146
  12. 2019Adversarial Training Can Hurt Generalizationcited by 132
  13. 2021Improving Question Answering Model Robustness with Synthetic Adversarial Data Generationcited by 79
  14. 2021How Should Pre-Trained Language Models Be Fine-Tuned Towards Adversarial Robustness?cited by 49
  15. 2018Effects of Degradations on Deep Neural Network Architecturescited by 48
  16. 2021Controlled Caption Generation for Images Through Adversarial Attackscited by 9
  17. 2023RoAST: Robustifying Language Models via Adversarial Perturbation with Selective Trainingcited by 3
  18. 2023Distribution-restrained Softmax Loss for the Model Robustnesscited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Survey of Adversarial Robustness in Multimodal Large Language Models
  2. 2024Multimodal Adversarial Defense for Vision-Language Models by Leveraging One-To-Many Relationships
  3. 2026Investigating Adversarial Robustness of Multi-modal Large Language Models
  4. 2023On the Robustness of Large Multimodal Models Against Image Adversarial Attacks
  5. 2024Sample-agnostic Adversarial Perturbation for Vision-Language Pre-training Models
  6. 2026Induced Numerical Instability: Hidden Costs in Multimodal Large Language Models
  7. 2023Exploring Transferability of Multimodal Adversarial Samples for Vision-Language Pre-training Models with Contrastive Learning
  8. 2025Robust-LLaVA: On the Effectiveness of Large-Scale Robust Image Encoders for Multi-modal Large Language Models