arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.12146cs.CLcs.AIcs.LG
收藏

通过元排序使弱LLM能够判断响应可靠性

Enabling Weak LLMs to Judge Response Reliability via Meta Ranking

Tsinghua University(清华大学);Weiyang College, Tsinghua University(清华大学未央书院);Institute of Intelligent Computing, Alibaba Group(阿里巴巴集团智能计算研究所);Institute for AI Industry Research (AIR), Tsinghua University(清华大学人工智能国际研究院)

Zijun Liu, Boqun Kou, Peng Li, Ming Yan, Ji Zhang, Fei Huang, Yang Liu

AI总结:

提出元排序(MR)方法,通过跨查询比较使弱LLM有效评估响应可靠性,在错误检测、模型级联和指令微调中显著提升效率与性能。

摘要:

尽管大型语言模型(LLMs)在广泛任务中表现出色,但仍存在可靠性问题。先前研究表明,像GPT-4-turbo这样的强LLM在评估LLM响应可靠性方面表现优异,但面临效率和本地部署问题。因此,为使弱LLM能有效评估LLM响应的可靠性,我们提出一种基于跨查询比较的新方法——元排序(Meta Ranking, MR)。与仅依赖LLM上下文学习能力的少样本方法不同,MR通过将目标查询-响应对与多个参考查询-响应对进行成对排序来评估可靠性。我们发现MR在LLM响应错误检测中极为有效,如Phi-2等弱LLM仅需五个参考样本即可超越GPT-3.5-turbo等强基线,并显著提升效率。我们进一步证明MR能增强强LLM在两个实际应用中的表现:模型级联和指令微调。在模型级联中,我们结合开源与闭源LLM,以更低成本实现媲美GPT-4-turbo的性能。在指令微调中,我们用MR进行迭代训练数据过滤,大幅缩短数据处理时间,使LLaMA-7B和Phi-2以更少训练token超越Alpaca-13B。这些结果凸显了MR在效率与有效性方面的巨大潜力。

英文摘要:

Despite the strong performance of large language models (LLMs) across a wide range of tasks, they still have reliability issues. Previous studies indicate that strong LLMs like GPT-4-turbo excel in evaluating the reliability of responses from LLMs, but face efficiency and local deployment issues. Thus, to enable weak LLMs to effectively assess the reliability of LLM responses, we propose a novel cross-query-comparison-based method called $\textit{Meta Ranking}$ (MR). Unlike previous few-shot methods that solely based on in-context learning capabilities in LLMs, MR assesses reliability by pairwisely ranking the target query-response pair with multiple reference query-response pairs. We found that MR is highly effective in error detection for LLM responses, where weak LLMs, such as Phi-2, could surpass strong baselines like GPT-3.5-turbo, requiring only five reference samples and significantly improving efficiency. We further demonstrate that MR can enhance strong LLMs' performance in two practical applications: model cascading and instruction tuning. In model cascading, we combine open- and closed-source LLMs to achieve performance comparable to GPT-4-turbo with lower costs. In instruction tuning, we use MR for iterative training data filtering, significantly reducing data processing time and enabling LLaMA-7B and Phi-2 to surpass Alpaca-13B with fewer training tokens. These results underscore the high potential of MR in both efficiency and effectiveness.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑