Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
Table 17 Ablation study of CoLog. Seq and Sem denote sequence and semantic, respectively. Also, CT, MHIA, MAL, and BL denote collaborative transformer, multi-head impressed attention, modality adaptation layer, and balancing layer.