arXiv 2401.165532024-08-27cs.CLcs.AI

SelectLLM：大语言模型能否选择重要指令进行标注？

SelectLLM: Can LLMs Select Important Instructions to Annotate?

University of Minnesota（明尼苏达大学）；Carnegie Mellon University（卡内基梅隆大学）

Ritik Sachin Parkar, Jaehyung Kim, Jong Inn Park, Dongyeop Kang

备注：: First Authors: Ritik Sachin Parkar and Jaehyung Kim | Second Author: Jong Inn Park | PI: Dongyeop Kang

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

SelectLLM框架利用大语言模型有效选择未标记指令，通过基于核心集的聚类增强多样性并提示LLM识别每簇中最有益的指令，在AlpacaEval2和MT-Bench上超越Alpagasus等方法，且兼容多种LLM及数据集。

摘要：

指令微调受益于大规模且多样化的数据集；然而，创建此类数据集需要高昂的人工标注成本。虽然由大语言模型（LLMs）生成的合成数据集部分解决了这一问题，但它们通常包含低质量数据。一种有效的解决方案是选择性地标注未标记指令，特别是考虑到从各种来源获取未标记指令或文本相对容易。然而，如何选择未标记指令尚未得到充分探索，尤其是在LLMs的背景下。因此，我们引入了SelectLLM，这是一种替代框架，利用LLMs的能力更有效地选择未标记指令。具体而言，SelectLLM包含两个关键步骤：基于核心集（Coreset）的未标记指令聚类以增强多样性，以及提示LLM在每个簇内识别最有益的指令。我们在AlpacaEval2和MT-Bench上评估了SelectLLM，展示了其超越Alpagasus等最先进方法的能力。此外，我们比较了SelectLLM与各种LLMs（如ChatGPT、LLaMA-3.1-70B和Gemma-2-27b）的性能和兼容性。SelectLLM的适应性和鲁棒性进一步得到验证，其能够在人工和合成数据集上均保持高性能。所有代码和数据均已公开（https://github.com/minnesotanlp/select-llm）。

英文摘要：

Instruction tuning benefits from large and diverse datasets; however, creating such datasets involves a high cost of human labeling. While synthetic datasets generated by large language models (LLMs) have partly solved this issue, they often contain low-quality data. One effective solution is selectively annotating unlabelled instructions, especially given the relative ease of acquiring unlabeled instructions or texts from various sources. However, how to select unlabelled instructions is not well-explored, especially in the context of LLMs. Therefore, we introduce SelectLLM, an alternative framework that leverages the capabilities of LLMs to select unlabeled instructions more effectively. Specifically, SelectLLM consists of two key steps: Coreset-based clustering of unlabelled instructions for enlarging diversity and prompting of LLM to identify the most beneficial instructions within each cluster. We evaluate SelectLLM on AlpacaEval2 and MT-Bench, demonstrating its ability to outperform state-of-the-art methods like Alpagasus. In addition, we compare the performance and compatibility of SelectLLM with various LLMs, such as ChatGPT, LLaMA-3.1-70B, and Gemma-2-27b. SelectLLM's adaptability and robustness are further evidenced by its ability to maintain high performance across both human and synthetic datasets. All code and data are publicly available (https://github.com/minnesotanlp/select-llm).

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。