Direct Preference Optimization (DPO) is a widely used offline preference optimization
algorithm that reparameterizes reward functions in reinforcement learning from human …

Direct Preference Optimisation (DPO) is effective at significantly improving the performance
of large language models (LLMs) on downstream tasks such as reasoning, summarisation …

Iterative preference optimization methods have recently been shown to perform well for
general instruction tuning tasks, but typically make little improvement on reasoning tasks …

While large language models (LLMs) like ChatGPT have shown impressive capabilities in
Natural Language Processing (NLP) tasks, a systematic investigation of their potential in this …

Insufficient modeling of human preferences within the reward model is a major obstacle for
leveraging human feedback to improve translation quality. Fortunately, quality estimation …

The development of large language models leads to the formation of a pre-train-then-align
paradigm, in which the model is typically pre-trained on a large text corpus and undergoes a …

Manually integrating data of diverse formats and languages is vital to many artificial
intelligence applications. However, the task itself remains challenging and time-consuming …

Direct Preference Optimization (DPO), which derives reward signals directly from pairwise
preference data, has shown its effectiveness on aligning Large Language Models (LLMs) …

Detoxifying multilingual Large Language Models (LLMs) has become crucial due to their
increasing global use. In this work, we explore zero-shot cross-lingual generalization of …

It is imperative for Large language models (LLMs) to follow instructions with elaborate
requirements (ie Complex Instructions Following). Yet, it remains under-explored how to …