Summary of Test-time Adaptation For Cross-modal Retrieval with Query Shift, by Haobin Li et al.
Test-time Adaptation for Cross-modal Retrieval with Query Shift
by Haobin Li, Peng Hu, Qianjun Zhang, Xi Peng, Xiting Liu, Mouxing Yang
First submitted to arxiv on: 21 Oct 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 proposes a novel method called Test-time adaptation for Cross-modal Retrieval (TCR) to address the query shift problem in cross-modal retrieval. Existing methods rely on the assumption that queries follow the same distribution as the source domain, but this assumption is often violated in real-world scenarios due to the complexity and diversity of queries. The proposed TCR method employs a novel module to refine query predictions and a joint objective to prevent query shift from disturbing the common space, allowing for online adaptation of cross-modal retrieval models with query shift. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper solves a problem that makes it hard for machines to find matching information when you ask them questions in different formats. This is called the query shift problem because your question can come from a different place than where the information came from. The paper proposes a new way to solve this problem, which they call Test-time adaptation for Cross-modal Retrieval (TCR). They tested it and showed that it works better than other methods. |
