Summary of Dynamic Subset Tuning: Expanding the Operational Range Of Parameter-efficient Training For Large Language Models, by Felix Stahlberg et al.
Dynamic Subset Tuning: Expanding the Operational Range of Parameter-Efficient Training for Large Language Models
by Felix Stahlberg, Jared Lichtarge, Shankar Kumar
First submitted to arxiv on: 13 Nov 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: 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 This novel parameter-efficient training (PET) method adapts large language models to downstream tasks by optimizing a dynamic subset of parameters. Unlike prior methods, the modified parameters evolve over the course of training, enabling good performance with many fewer parameters than extant methods. Our approach seamlessly scales the subset size across an arbitrary proportion of the total model size, outperforming prompt tuning and LoRA in most cases on various NLP tasks (MT, QA, GSM8K, SuperGLUE) for a given parameter budget. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a new way to train big language models. Instead of using all the parameters, it only uses a small part that changes during training. This helps the model perform well with fewer parameters than other methods. The method can adjust the number of used parameters to fit different tasks and models. It works as well or better than existing methods on various tasks like machine translation, question answering, and text classification. |
Keywords
» Artificial intelligence » Lora » Nlp » Parameter efficient » Prompt » Question answering » Text classification » Translation
