Large language models (LLMs) are gaining increasing popularity in both academia and
industry, owing to their unprecedented performance in various applications. As LLMs …

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

Abstract Large Language Models (LLMs), despite their recent impressive accomplishments,
are notably cost-prohibitive to deploy, particularly for applications involving long-content …

Large Language Models (LLMs) have unlocked new capabilities and applications; however,
evaluating the alignment with human preferences still poses significant challenges. To …

We introduce the Yi model family, a series of language and multimodal models that
demonstrate strong multi-dimensional capabilities. The Yi model family is based on 6B and …

In this work we explore recent advances in instruction-tuning language models on a range of
open instruction-following datasets. Despite recent claims that open models can be on par …

Despite the advancements of open-source large language models (LLMs), eg, LLaMA, they
remain significantly limited in tool-use capabilities, ie, using external tools (APIs) to fulfill …

In the rapidly evolving domain of Natural Language Generation (NLG) evaluation,
introducing Large Language Models (LLMs) has opened new avenues for assessing …