Generating a Paracosm for Training-Free Zero-Shot Composed Image Retrieval
Tong Wang, Yunhan Zhao, Shu Kong · 2026
Read Generating a Paracosm for Training-Free Zero-Shot Composed Image Retrieval by Tong Wang, Yunhan Zhao, Shu Kong (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-00813.txt.
Description
Tong Wang and colleagues generate the mental image itself for composed image retrieval, instead of describing it in words. A composed retrieval query pairs a reference image with text saying how to change it. The target mental image exists only implicitly. Prior zero-shot work has a large multimodal model write a textual description for matching; they search directly with the generated image instead. Because the synthetic mental image differs from real images, they make a synthetic counterpart for each database image to ease matching.
Catalog record
| CREATOR | Tong Wang, Yunhan Zhao, Shu Kong |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 64320 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602.00813