Research map / Adversarial robustness and security / Attacks on text and graphs
Defending Multimodal Fusion Models Against Single-Source Adversaries
Karren Yang, Wan-Yi Lin, Manash Barman, Filipe Condessa, Zico Kolter
Check your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use odd-one-out robust fusion with adversarial training of the fusion modules | Multimodal models · EPIC-Kitchens, single-source attacks | Top-1 action accuracy: 7.8-25.2% | Improved |
| Use odd-one-out robust fusion with adversarial training of the fusion modules | Multimodal models · KITTI, single-source attacks | Average precision: 19.7-48.2% | Improved |
| Use odd-one-out robust fusion with adversarial training of the fusion modules | Multimodal models · CMU-MOSI, single-source attacks | Accuracy: 1.6-6.7% | Improved |
| Use standard concatenation or mean fusion trained on clean data | Multimodal models · EPIC-Kitchens and KITTI, single-source attacks | Accuracy or average precision | Hurt |
Closest papers by meaning
- On the Robustness of Large Multimodal Models Against Image Adversarial Attacks (CVPR 2024 · similarity 0.89)
- FusionAudit: Pathway-Conditioned Robustness Auditing for Native Multimodal Models (NeurIPS 2026 · similarity 0.86)
- A Closer Look at Multimodal Representation Collapse (ICML 2025 · similarity 0.86)
- Quantifying and Enhancing Multi-modal Robustness with Modality Preference (ICLR 2024 · similarity 0.86)
- Fusion or Confusion? Multimodal Complexity Is Not All You Need (NeurIPS 2026 · similarity 0.85)
- Fusion Is Not Enough: Single Modal Attacks on Fusion Models for 3D Object Detection (ICLR 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
