SparsePO: Controlling Preference Alignment of LLMs via Sparse Token Masks
"偏好对齐" 相关文章
Fine-tuning Large Language Models with Limited Data: A Survey and Practical Guide
Margin Adaptive DPO: Leveraging Reward Model for Granular Control in Preference Optimization
A Contextual Quality Reward Model for Reliable and Efficient Best-of-N Sampling
Pluralistic Off-policy Evaluation and Alignment
OmniDPO: A Preference Optimization Framework to Address Omni-Modal Hallucination
我们都错怪GPT-5了，路由统一算力，免费用户也能创造收益
MixGRPO: Unlocking Flow-based GRPO Efficiency with Mixed ODE-SDE
PITA: Preference-Guided Inference-Time Alignment for LLM Post-Training
PIPA: Preference Alignment as Prior-Informed Statistical Estimation
Koel-TTS: Enhancing LLM based Speech Generation with Preference Alignment and Classifier Free Guidance
扩散语言模型扛把子LLaDA迎来新版本，数学、代码、对齐能力均提升
ICML 2025 | RLHF太贵太慢？TPO即时对齐新方案，一句话指令搞定偏好优化
把RLHF带给VLA模型！通过偏好对齐来优化机器人策略，代码已开源
Meet GRAPE: A Plug-and-Play Algorithm to Generalize Robot Policies via Preference Alignment
多图场景用DPO对齐！上海AI实验室等提出新方法，无需人工标注
多图场景用DPO对齐，上海AI实验室等提出新方法，无需人工标注
