讲座｜蒋联江：AI赋能英语教育中的多模态评估创新

This talk aims to prepare English language teachers to use artificial intelligence (AI) to design and implement multimodal assessment tasks in English language education. As traditional assessment is increasingly vulnerable in the era of GenAI, multimodal assessment is merited as a responsive assessment innovation. In general, multimodal assessment refers to evaluation methods involving two or more modes of communication such as image, writing, speech, video, soundtrack and 3D objects to represent learning and knowledge. Multimodal assessment is considered as a potential way to make English learning and teaching responsive to increasing diversity in contemporary literacy practices, students’ linguistic and cultural backgrounds, and digital media and communication channels. However, designing multimodal assessment tasks remains a challenge for English teachers given its absence in traditional language teacher education programs. The emergence of AI technologies offers both opportunities and challenges for multimodal assessment task design. This talk begins with an evidence-based framework for multimodal assessment task design. The role and potential use of AI in each stage of the framework will be specified. Hands-on activities of using various AI tools in multimodal assessment task design will also be provided, including AI-supported multimodal resource generation, combination, and rubric writing. The workshop participants are encouraged to install the following AI and AI-embedded tools on their devices, including Canva, ElevenLabs, DALLE-2, and DeepSeek, for the hands-on activities at the workshop.

蒋联江教授是香港大学教育学院语文研究与教育学部助理教授，香港大学语言政策与实践研究联盟（CRLPP）副主任。他重点关注数字化多模态写作，数码素养，多模态评估与反馈。他主持多项香港特别行政区研究资助局和教育局项目，研究经费逾300万港元。他是TESOL Quarterly研究传播编辑，The JALT CALL Journal (Scopus-Q1)副主编，也是Computer Assisted Language Learning, TESOL Quarterly, Journal of Second Language Writing, Assessing Writing, Linguistics and Education等期刊编委会成员。他入选科睿唯安全球前1%高被引研究者以及美国斯坦福大学和爱思唯尔全球前2%顶尖科学家榜单。