Summary of Learning From True-false Labels Via Multi-modal Prompt Retrieving, by Zhongnian Li et al.
Learning from True-False Labels via Multi-modal Prompt Retrieving
by Zhongnian Li, Jinghao Xu, Peng Ying, Meng Wei, Tongfeng Sun, Xinzheng Xu
First submitted to arxiv on: 24 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computer Vision and Pattern Recognition (cs.CV)
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 paper proposes a novel weakly supervised learning approach, True-False Labels (TFLs), which leverages pre-trained vision-language models (VLMs) to generate reliable labels. The TFL setting indicates whether an instance belongs to a randomly sampled label from the candidate set. The authors derive a risk-consistent estimator to utilize conditional probability distribution information and introduce a convolutional-based Multi-modal Prompt Retrieving (MRP) method to bridge the gap between VLM knowledge and target learning tasks. Experimental results demonstrate the effectiveness of the proposed approach. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about finding ways to label things without needing too much human help. Right now, using pre-trained models isn’t very good at labeling things accurately. The authors came up with a new way called True-False Labels (TFLs) that uses these pre-trained models better. They also developed a way to use the knowledge from these models to do tasks like image classification and object detection. This can make it easier and cheaper to train AI models. |
Keywords
» Artificial intelligence » Image classification » Multi modal » Object detection » Probability » Prompt » Supervised
