AI RESEARCH
ADaFuSE: Adaptive Diffusion-generated Image and Text Fusion for Interactive Text-to-Image Retrieval
arXiv CS.CV
•
ArXi:2603.21886v1 Announce Type: cross Recent advances in interactive text-to-image retrieval (I-TIR) use diffusion models to bridge the modality gap between the textual information need and the images to be searched, resulting in increased effectiveness. However, existing frameworks fuse multi-modal views of user feedback by simple embedding addition. In this work, we show that this static and undifferentiated fusion indiscriminately incorporates generative noise produced by the diffusion model, leading to performance degradation for up to 55.62% samples.