 On the Limited Generalization Capability of the Implicit Reward Model
Induced by Direct Preference OptimizationConference on Empirical Methods in Natural Language Processing (EMNLP), 2024 |
 Regularizing Hidden States Enables Learning Generalizable Reward Model
for LLMsNeural Information Processing Systems (NeurIPS), 2024 |
 Self-Play Fine-Tuning Converts Weak Language Models to Strong Language
ModelsInternational Conference on Machine Learning (ICML), 2024 |
 On Diversified Preferences of Large Language Model AlignmentConference on Empirical Methods in Natural Language Processing (EMNLP), 2023 |
