Semantic Token Reweighting for Interpretable and Controllable Text Embeddings in CLIP
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910652598059008 |
|---|---|
| author | Kim, Eunji Shim, Kyuhong Chang, Simyung Yoon, Sungroh |
| author_facet | Kim, Eunji Shim, Kyuhong Chang, Simyung Yoon, Sungroh |
| contents | A text encoder within Vision-Language Models (VLMs) like CLIP plays a crucial role in translating textual input into an embedding space shared with images, thereby facilitating the interpretative analysis of vision tasks through natural language. Despite the varying significance of different textual elements within a sentence depending on the context, efforts to account for variation of importance in constructing text embeddings have been lacking. We propose a framework of Semantic Token Reweighting to build Interpretable text embeddings (SToRI), which incorporates controllability as well. SToRI refines the text encoding process in CLIP by differentially weighting semantic elements based on contextual importance, enabling finer control over emphasis responsive to data-driven insights and user preferences. The efficacy of SToRI is demonstrated through comprehensive experiments on few-shot image classification and image retrieval tailored to user preferences. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_08469 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Semantic Token Reweighting for Interpretable and Controllable Text Embeddings in CLIP Kim, Eunji Shim, Kyuhong Chang, Simyung Yoon, Sungroh Machine Learning Computation and Language Computer Vision and Pattern Recognition A text encoder within Vision-Language Models (VLMs) like CLIP plays a crucial role in translating textual input into an embedding space shared with images, thereby facilitating the interpretative analysis of vision tasks through natural language. Despite the varying significance of different textual elements within a sentence depending on the context, efforts to account for variation of importance in constructing text embeddings have been lacking. We propose a framework of Semantic Token Reweighting to build Interpretable text embeddings (SToRI), which incorporates controllability as well. SToRI refines the text encoding process in CLIP by differentially weighting semantic elements based on contextual importance, enabling finer control over emphasis responsive to data-driven insights and user preferences. The efficacy of SToRI is demonstrated through comprehensive experiments on few-shot image classification and image retrieval tailored to user preferences. |
| title | Semantic Token Reweighting for Interpretable and Controllable Text Embeddings in CLIP |
| topic | Machine Learning Computation and Language Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2410.08469 |