Summary of Conifer: Improving Complex Constrained Instruction-following Ability Of Large Language Models, by Haoran Sun and Lixin Liu and Junjie Li and Fengyu Wang and Baohua Dong and Ran Lin and Ruohui Huang
Conifer: Improving Complex Constrained Instruction-Following Ability of Large Language Models
by Haoran Sun, Lixin Liu, Junjie Li, Fengyu Wang, Baohua Dong, Ran Lin, Ruohui Huang
First submitted to arxiv on: 3 Apr 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper introduces Conifer, a novel instruction tuning dataset designed to enhance large language models (LLMs) in following multi-level instructions with complex constraints. The authors utilize GPT-4 and refine the dataset through LLM-driven processes to ensure high quality. They also propose a progressive learning scheme emphasizing easy-to-hard progression and process feedback. Models trained on Conifer exhibit significant improvements in instruction-following abilities, especially for complex constraint instructions. The 7B model outperforms state-of-the-art open-source 7B models on certain benchmarks and even surpasses larger models (10 times) on specific metrics. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Conifer is a new dataset that helps big language models follow instructions better. Right now, these models struggle when given complex instructions with many rules. The researchers created Conifer to fix this problem by making the models learn from easy instructions first and then harder ones. They also made sure the data was high quality by having the models help refine it themselves. This new approach makes a big difference in how well the models can follow instructions, especially when there are many complex rules involved. |
Keywords
* Artificial intelligence * Gpt * Instruction tuning
