The ALVR workshop at ACL 2026 accepts 29 papers on multimodal learning and large vision-language models · Digg