CURATRON: Complete and Robust Preference Data for Rigorous Alignment of Large Language Models
"偏好学习" 相关文章
Preference Learning with Response Time: Robust Losses and Guarantees
Representer Theorems for Metric and Preference Learning: Geometric Insights and Algorithms
Preference-based Reinforcement Learning beyond Pairwise Comparisons: Benefits of Multiple Options
Direct Preference Optimization with Unobserved Preference Heterogeneity: The Necessity of Ternary Preferences
RealDPO: Real or Not Real, that is the Preference
Beyond Correctness: Evaluating Subjective Writing Preferences Across Cultures
Beyond Correctness: Evaluating Subjective Writing Preferences Across Cultures
Repairing Reward Functions with Human Feedback to Mitigate Reward Hacking
Repairing Reward Functions with Human Feedback to Mitigate Reward Hacking
大模型不会用工具？人大Tool-Light：不存在的！
Reward Model Perspectives: Whose Opinions Do Reward Models Reward?
Solving the Granularity Mismatch: Hierarchical Preference Learning for Long-Horizon LLM Agents
Annotation-Efficient Language Model Alignment via Diverse and Representative Response Texts
Learning to Plan with Personalized Preferences
Temporal Preference Optimization for Long-Form Video Understanding
Temporal Self-Rewarding Language Models: Decoupling Chosen-Rejected via Past-Future
将偏好学习引入模型训练，北大李戈团队新框架，可显著提升代码准确性与执行效率
Evaluation is All You Need！首个开源多模态大模型通用评测器LLaVA-Critic
