Abstract
We introduce ChatGLM, an evolving family of large language models that we have been developing over time. This report primarily focuses on the GLM-4 language series, which includes GLM-4, GLM-4-Air, and GLM-4-9B. They represent our most capable models that are trained with all the insights and lessons gained from the preceding three generations of ChatGLM. To date, the GLM-4 models are pre-trained on ten trillions of tokens mostly in Chinese and English, along with a small set of corpus from 24 languages, and aligned primarily for Chinese and English usage. The high-quality alignment is achieved via a multi-stage post-training process, which involves supervised fine-tuning and learning from human feedback. Evaluations show that GLM-4 1) closely rivals or outperforms GPT-4 in terms of general metrics such as MMLU, GSM8K, MATH, BBH, GPQA, and HumanEval, 2) gets close to GPT-4-Turbo in instruction following as measured by IFEval, 3) matches GPT-4 Turbo (128K) and Claude 3 for long context tasks, and 4) outperforms GPT-4 in Chinese alignments as measured by AlignBench. The GLM-4 All Tools model is further aligned to understand user intent and autonomously decide when and which tool(s) touse -- including web browser, Python interpreter, text-to-image model, and user-defined functions -- to effectively complete complex tasks. In practical applications, it matches and even surpasses GPT-4 All Tools in tasks like accessing online information via web browsing and solving math problems using Python interpreter. Over the course, we have open-sourced a series of models, including ChatGLM-6B (three generations), GLM-4-9B (128K, 1M), GLM-4V-9B, WebGLM, and CodeGeeX, attracting over 10 million downloads on Hugging face in the year 2023 alone. The open models can be accessed through this https URL and this https URL.
Abstract (translated)
我们介绍了一个不断演变的家族大型语言模型 ChatGLM，这是我们近年来开发的一系列活动。这份报告主要关注 GLM-4 语言系列，包括 GLM-4、GLM-4-Air 和 GLM-4-9B。它们是我们通过前三代 ChatGLM 获得的最具能力的模型。到目前为止，GLM-4 模型都在中文和英语等语言中进行预训练，配备了一个小规模的语料库（24 种语言），主要针对中文和英语使用进行对齐。高质量的对齐是通过多阶段的后训练过程实现的，包括有监督的微调和对人类反馈的学习。评估显示，GLM-4 1）与 GPT-4 在诸如 MMLU、GSM8K、MATH、BBH、GPQA 和 HumanEval 等指标上紧密竞争或超越，2）在指令跟随方面接近 GPT-4-Turbo，3）在长文本任务中与 GPT-4 Turbo（128K）和 Claude 3 相当，4）在中文对齐方面优于 GPT-4。GLM-4 All Tools 模型进一步对齐以理解用户意图并自主决定何时以及使用哪些工具（包括浏览器、Python 解释器、图像-文本模型和用户定义函数）来有效地完成复杂任务。在实际应用中，它在通过网页浏览访问在线信息和解数学问题方面甚至超过了 GPT-4 All Tools。在模型开源方面，我们开源了一系列模型，包括 ChatGLM-6B（三代），GLM-4-9B（128K，1M），GLM-4V-9B，WebGLM 和 CodeGeeX，在 2023 年仅通过 Hugging Face 下载量超过 1000 万次。这些开源模型可以通过这个 https://url 和这个 https://url 访问。
URL
https://arxiv.org/abs/2406.12793
