Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation
CoRR(2024)
摘要
Moderate-sized large language models (LLMs) – those with 7B or 13B
parameters – exhibit promising machine translation (MT) performance. However,
even the top-performing 13B LLM-based translation models, like ALMA, does not
match the performance of state-of-the-art conventional encoder-decoder
translation models or larger-scale LLMs such as GPT-4. In this study, we bridge
this performance gap. We first assess the shortcomings of supervised
fine-tuning for LLMs in the MT task, emphasizing the quality issues present in
the reference data, despite being human-generated. Then, in contrast to SFT
which mimics reference translations, we introduce Contrastive Preference
Optimization (CPO), a novel approach that trains models to avoid generating
adequate but not perfect translations. Applying CPO to ALMA models with only
22K parallel sentences and 12M parameters yields significant improvements. The
resulting model, called ALMA-R, can match or exceed the performance of the WMT
competition winners and GPT-4 on WMT'21, WMT'22 and WMT'23 test datasets.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要