Towards Harmless Multimodal Assistants with Blind Preference Optimization
Open Access
- 27 October 2025
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 4992-5000
Abstract
No abstract availableAll Related Versions
Funding Information
- Research Grants Council of Hong Kong (15207122,15213323,15205325)
- The Hong Kong Polytechnic University (BDWP)
This publication has 13 references indexed in Scilit:
- Revolutionizing Text-to-Image Retrieval as Autoregressive Token-to-Voken Generation2025
- SPA-VL: A Comprehensive Safety Preference Alignment Dataset for Vision Language Models2025
- Improved Baselines with Visual Instruction Tuning2024
- DRESS : Instructing Large Vision-Language Models to Align and Interact with Humans via Natural Language Feedback2024
- RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-Grained Correctional Human Feedback2024
- Visual Adversarial Examples Jailbreak Aligned Large Language ModelsProceedings of the AAAI Conference on Artificial Intelligence, 2024
- Generative Bias for Robust Visual Question Answering2023
- Cross-modal Moment Localization in Videos2018
- Attentive Moment Retrieval in Videos2018
- Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering2017