Do VLMs Share Safety Neurons Across Modalities?
cs.LG
Submitted: 2026-08-31
Updated: 2026-08-31
Comments: Accepted to EMNLP 2026 (Main Conference)
Project page: https://jiaxuan-li.github.io/vlm-safety-neuron
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Qwen3-VL Technical Report
- Qwen2.5-VL Technical Report
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- Gemma 4 Technical Report
- GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning
- GPT-4o System Card
- InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency
- The Hydra Effect: Emergent Self-repair in Language Model Computations
- Representation Engineering: A Top-Down Approach to AI Transparency
- Universal and Transferable Adversarial Attacks on Aligned Language Models
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks