1. Weaver: Foundation Models for Creative Writing
- Author
-
Wang, Tiannan, Chen, Jiamin, Jia, Qingrui, Wang, Shuai, Fang, Ruoyu, Wang, Huilin, Gao, Zhaowei, Xie, Chunzhao, Xu, Chuou, Dai, Jihong, Liu, Yibin, Wu, Jialong, Ding, Shengwei, Li, Long, Huang, Zhiwei, Deng, Xinle, Yu, Teng, Ma, Gangan, Xiao, Han, Chen, Zixin, Xiang, Danjun, Wang, Yunxia, Zhu, Yuanyuan, Xiao, Yi, Wang, Jing, Wang, Yiru, Ding, Siran, Huang, Jiayang, Xu, Jiayi, Tayier, Yilihamu, Hu, Zhenyu, Gao, Yuan, Zheng, Chengfeng, Ye, Yueshu, Li, Yihang, Wan, Lei, Jiang, Xinyue, Wang, Yujie, Cheng, Siyu, Song, Zhule, Tang, Xiangru, Xu, Xiaohua, Zhang, Ningyu, Chen, Huajun, Jiang, Yuchen Eleanor, and Zhou, Wangchunshu
- Subjects
Computer Science - Computation and Language ,Computer Science - Artificial Intelligence ,Computer Science - Machine Learning - Abstract
This work introduces Weaver, our first family of large language models (LLMs) dedicated to content creation. Weaver is pre-trained on a carefully selected corpus that focuses on improving the writing capabilities of large language models. We then fine-tune Weaver for creative and professional writing purposes and align it to the preference of professional writers using a suit of novel methods for instruction data synthesis and LLM alignment, making it able to produce more human-like texts and follow more diverse instructions for content creation. The Weaver family consists of models of Weaver Mini (1.8B), Weaver Base (6B), Weaver Pro (14B), and Weaver Ultra (34B) sizes, suitable for different applications and can be dynamically dispatched by a routing agent according to query complexity to balance response quality and computation cost. Evaluation on a carefully curated benchmark for assessing the writing capabilities of LLMs shows Weaver models of all sizes outperform generalist LLMs several times larger than them. Notably, our most-capable Weaver Ultra model surpasses GPT-4, a state-of-the-art generalist LLM, on various writing scenarios, demonstrating the advantage of training specialized LLMs for writing purposes. Moreover, Weaver natively supports retrieval-augmented generation (RAG) and function calling (tool usage). We present various use cases of these abilities for improving AI-assisted writing systems, including integration of external knowledge bases, tools, or APIs, and providing personalized writing assistance. Furthermore, we discuss and summarize a guideline and best practices for pre-training and fine-tuning domain-specific LLMs.
- Published
- 2024