較小的語言模型更適合進化指導者。

Smaller Language Models Are Better Instruction Evolvers

December 15, 2024

作者: Tingfeng Hui, Lulu Zhao, Guanting Dong, Yaqi Zhang, Hua Zhou, Sen Su

cs.AI

摘要

指令調整廣泛應用於發揮大型語言模型的完整潛力。值得注意的是，複雜且多樣化的指令具有重要意義，因為它們可以有效地使模型與各種下游任務保持一致。然而，目前構建大規模指令的方法主要偏好強大的模型，如GPT-4或具有超過 700 億參數的模型，這是基於這樣一種經驗假設：這樣更大的語言模型（LLMs）本質上具有增強的能力。在本研究中，我們質疑這種普遍的假設，並深入探討在指令演進的背景下，較小語言模型（SLMs）的潛力。通過對指令演進的三種情境進行廣泛實驗，我們發現較小語言模型（SLMs）可以比LLMs合成更有效的指令。進一步分析表明，在指令演進過程中，SLMs擁有更廣泛的輸出空間，從而產生更複雜和多樣化的變體。我們還觀察到現有的指標未能專注於指令的影響。因此，我們提出指令複雜感知 IFD（IC-IFD），它在原始 IFD 分數中引入指令複雜度，以更準確地評估指令數據的有效性。我們的原始碼可在以下鏈接找到：https://github.com/HypherX/Evolution-Analysis {https://github.com/HypherX/Evolution-Analysis}

English

Instruction tuning has been widely used to unleash the complete potential of large language models. Notably, complex and diverse instructions are of significant importance as they can effectively align models with various downstream tasks. However, current approaches to constructing large-scale instructions predominantly favour powerful models such as GPT-4 or those with over 70 billion parameters, under the empirical presumption that such larger language models (LLMs) inherently possess enhanced capabilities. In this study, we question this prevalent assumption and conduct an in-depth exploration into the potential of smaller language models (SLMs) in the context of instruction evolution. Extensive experiments across three scenarios of instruction evolution reveal that smaller language models (SLMs) can synthesize more effective instructions than LLMs. Further analysis demonstrates that SLMs possess a broader output space during instruction evolution, resulting in more complex and diverse variants. We also observe that the existing metrics fail to focus on the impact of the instructions. Thus, we propose Instruction Complex-Aware IFD (IC-IFD), which introduces instruction complexity in the original IFD score to evaluate the effectiveness of instruction data more accurately. Our source code is available at: https://github.com/HypherX/Evolution-Analysis{https://github.com/HypherX/Evolution-Analysis}