Summary of Controllable Prompt Tuning For Balancing Group Distributional Robustness, by Hoang Phan and Andrew Gordon Wilson and Qi Lei
Controllable Prompt Tuning For Balancing Group Distributional Robustness
by Hoang Phan, Andrew Gordon Wilson, Qi Lei
First submitted to arxiv on: 5 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 an optimization scheme to improve the performance of models trained on multi-group or domain datasets under distribution shifts. The goal is to achieve good performance across all groups without sacrificing performance on any one group. The authors propose Controllable Prompt Tuning (CPT), which combines their approach with prompt-tuning techniques, requiring only 0.4% tunable parameters. CPT achieves state-of-the-art results on spurious correlation benchmarks for transformer and non-transformer architectures, as well as unimodal and multimodal data. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps models understand different groups or domains better by finding a good solution for all without sacrificing performance on any one group. The authors create a way to do this using something called Controllable Prompt Tuning (CPT). CPT is a combination of their approach and another technique that only needs a small number of parameters to be adjusted. This makes it efficient and effective. |
Keywords
* Artificial intelligence * Optimization * Prompt * Transformer
