CoachLM: Automatic Instruction Revisions Improve the Data Quality in LLM Instruction Tuning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911806353571840 |
|---|---|
| author | Liu, Yilun Tao, Shimin Zhao, Xiaofeng Zhu, Ming Ma, Wenbing Zhu, Junhao Su, Chang Hou, Yutai Zhang, Miao Zhang, Min Ma, Hongxia Zhang, Li Yang, Hao Jiang, Yanfei |
| author_facet | Liu, Yilun Tao, Shimin Zhao, Xiaofeng Zhu, Ming Ma, Wenbing Zhu, Junhao Su, Chang Hou, Yutai Zhang, Miao Zhang, Min Ma, Hongxia Zhang, Li Yang, Hao Jiang, Yanfei |
| contents | Instruction tuning is crucial for enabling Language Learning Models (LLMs) in responding to human instructions. The quality of instruction pairs used for tuning greatly affects the performance of LLMs. However, the manual creation of high-quality instruction datasets is costly, leading to the adoption of automatic generation of instruction pairs by LLMs as a popular alternative. To ensure the high quality of LLM-generated instruction datasets, several approaches have been proposed. Nevertheless, existing methods either compromise dataset integrity by filtering a large proportion of samples, or are unsuitable for industrial applications. In this paper, instead of discarding low-quality samples, we propose CoachLM, a novel approach to enhance the quality of instruction datasets through automatic revisions on samples in the dataset. CoachLM is trained from the samples revised by human experts and significantly increases the proportion of high-quality samples in the dataset from 17.7% to 78.9%. The effectiveness of CoachLM is further assessed on various real-world instruction test sets. The results show that CoachLM improves the instruction-following capabilities of the instruction-tuned LLM by an average of 29.9%, which even surpasses larger LLMs with nearly twice the number of parameters. Furthermore, CoachLM is successfully deployed in a data management system for LLMs at Huawei, resulting in an efficiency improvement of up to 20% in the cleaning of 40k real-world instruction pairs. We release various assets of CoachLM, including the training data, code and test set (https://github.com/lunyiliu/CoachLM). |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2311_13246 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | CoachLM: Automatic Instruction Revisions Improve the Data Quality in LLM Instruction Tuning Liu, Yilun Tao, Shimin Zhao, Xiaofeng Zhu, Ming Ma, Wenbing Zhu, Junhao Su, Chang Hou, Yutai Zhang, Miao Zhang, Min Ma, Hongxia Zhang, Li Yang, Hao Jiang, Yanfei Computation and Language Instruction tuning is crucial for enabling Language Learning Models (LLMs) in responding to human instructions. The quality of instruction pairs used for tuning greatly affects the performance of LLMs. However, the manual creation of high-quality instruction datasets is costly, leading to the adoption of automatic generation of instruction pairs by LLMs as a popular alternative. To ensure the high quality of LLM-generated instruction datasets, several approaches have been proposed. Nevertheless, existing methods either compromise dataset integrity by filtering a large proportion of samples, or are unsuitable for industrial applications. In this paper, instead of discarding low-quality samples, we propose CoachLM, a novel approach to enhance the quality of instruction datasets through automatic revisions on samples in the dataset. CoachLM is trained from the samples revised by human experts and significantly increases the proportion of high-quality samples in the dataset from 17.7% to 78.9%. The effectiveness of CoachLM is further assessed on various real-world instruction test sets. The results show that CoachLM improves the instruction-following capabilities of the instruction-tuned LLM by an average of 29.9%, which even surpasses larger LLMs with nearly twice the number of parameters. Furthermore, CoachLM is successfully deployed in a data management system for LLMs at Huawei, resulting in an efficiency improvement of up to 20% in the cleaning of 40k real-world instruction pairs. We release various assets of CoachLM, including the training data, code and test set (https://github.com/lunyiliu/CoachLM). |
| title | CoachLM: Automatic Instruction Revisions Improve the Data Quality in LLM Instruction Tuning |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2311.13246 |