Summary of Packing Analysis: Packing Is More Appropriate For Large Models or Datasets in Supervised Fine-tuning, by Shuhe Wang et al.
Packing Analysis: Packing Is More Appropriate for Large Models or Datasets in Supervised Fine-tuning
by Shuhe Wang, Guoyin Wang, Yizhong Wang, Jiwei Li, Eduard Hovy, Chen Guo
First submitted to arxiv on: 10 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
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 This research paper investigates the effectiveness of packing, a technique that optimizes hardware resource efficiency during pre-training, when applied to supervised fine-tuning (SFT) stages. The authors analyze whether packing can enhance training efficiency while maintaining performance, determine suitable model and dataset sizes for fine-tuning with packing, and examine how unrelated or related training samples affect the model’s behavior. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Packing is a technique that helps make computer learning more efficient by combining different types of information to fit a model’s maximum input size. This paper looks at whether using this technique during the fine-tuning stage can make the process faster while still producing good results. It also tries to figure out what kinds of models and datasets work best with this method and how it affects how well the model learns from different types of information. |
Keywords
» Artificial intelligence » Fine tuning » Supervised
