Defending LVLMs Against Vision Attacks Through Partial-Perception Supervision

Qi Zhou, Dongxia Wang, Tianlin Li, Yun Lin, Yang Liu +2 more

2/14/2026

No official code found. Best implementation: tools-only/DPS (5 stars, last commit 2025-07-14).

Semantic Scholar

Code Implementations(1)

tools-only/DPSCommunity65%

Code for "Defending LVLMs Against Vision Attacks through Partial-Perception Supervision"

50Python, Jupyter NotebookApr 1, 20251 years ago

Similar Papers

Scaling Sparse Feature Circuits For Studying In-Context Learning

Controllable Blur Data Augmentation Using 3D-Aware Motion Estimation

Training A Secure Model Against Data-Free Model Extraction

100% match1 repo2026

Online Robust Reinforcement Learning Through Monte-Carlo Planning

Task Descriptors Help Transformers Learn Linear Models In-Context

Exposure Bracketing Is All You Need For A High-Quality Image

100% match1 repo1262026

Cite this paper

@article{zhou2026defending,
  title  = {Defending LVLMs Against Vision Attacks Through Partial-Perception Supervision},
  author = {Qi Zhou and Dongxia Wang and Tianlin Li and Yun Lin and Yang Liu and Jin Song Dong and Qing Guo},
  year   = {2026},
  url    = {https://api.semanticscholar.org/CorpusID:571b4636613243f690eb73bf86634faf60bfa5d6},
  journal = {ICML 2025 2025}
}

Add a badge to your README

Shows whether official code is linked for this paper.

[![Code of Paper](https://codeofpaper.com/badge/571b4636613243f690eb73bf86634faf60bfa5d6.svg)](https://codeofpaper.com/paper/571b4636613243f690eb73bf86634faf60bfa5d6)

Discussion