preference pairs

6 papers

• DP²O-SR: Direct Perceptual Preference Optimization for Real-World Image Super-Resolution
• Quantile Reward Policy Optimization: Alignment with Pointwise Regression and Exact Partition Functions
• Self-Supervised Direct Preference Optimization for Text-to-Image Diffusion Models
• Sparta Alignment: Collectively Aligning Multiple Language Models through Combat
• Systematic Reward Gap Optimization for Mitigating VLM Hallucinations
• VPO: Reasoning Preferences Optimization Based on $\mathcal{V}$-Usable Information