Summary of Parameter-efficient Fine-tuning Of State Space Models, by Kevin Galim et al.
Parameter-Efficient Fine-Tuning of State Space Models
by Kevin Galim, Wonjun Kang, Yuchen Zeng, Hyung Il Koo, Kangwook Lee
First submitted to arxiv on: 11 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: 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 The paper investigates the application of parameter-efficient fine-tuning (PEFT) methods to Deep State Space Models (SSMs), specifically Mamba models. It evaluates existing PEFT methods on SSM-based models and identifies LoRA as a consistently high-performing method. However, LoRA is ineffective for SSM modules, highlighting the need for a specialized tuning approach. The authors propose Sparse Dimension Tuning (SDT) for SSMs, which combines with LoRA for linear projection matrices to achieve state-of-the-art performance. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Deep State Space Models are powerful tools for language modeling that offer high performance and scalability. But what happens when you try to fine-tune these models? The paper looks at how different methods work on these models and finds that one method, LoRA, is especially good. However, it’s not perfect and actually doesn’t work well with certain parts of the model. To fix this, the authors come up with a new way to tune the model called SDT. When they use SDT together with LoRA, they get amazing results. |
Keywords
» Artificial intelligence » Fine tuning » Lora » Parameter efficient
