{"type":"ScholarlyArticle","title":"ChatSpot: Bootstrapping Multimodal LLMs via Precise Referring Instruction Tuning","authors":[{"name":"Liang Zhao","slug":"liang-zhao","given":"Liang","family":"Zhao","affiliation":""},{"name":"En Yu","slug":"en-yu","given":"En","family":"Yu","affiliation":""},{"name":"Zheng Ge","slug":"zheng-ge","given":"Zheng","family":"Ge","affiliation":""},{"name":"Jinrong Yang","slug":"jinrong-yang","given":"Jinrong","family":"Yang","affiliation":""},{"name":"Haoran Wei","slug":"haoran-wei","given":"Haoran","family":"Wei","affiliation":""},{"name":"Hongyu Zhou","slug":"hongyu-zhou","given":"Hongyu","family":"Zhou","affiliation":""},{"name":"Jianjian Sun","slug":"jianjian-sun","given":"Jianjian","family":"Sun","affiliation":""},{"name":"Yuang Peng","slug":"yuang-peng","given":"Yuang","family":"Peng","affiliation":""},{"name":"Runpei Dong","slug":"runpei-dong","given":"Runpei","family":"Dong","affiliation":""},{"name":"Chunrui Han","slug":"chunrui-han","given":"Chunrui","family":"Han","affiliation":""},{"name":"Xiangyu Zhang","slug":"xiangyu-zhang","given":"Xiangyu","family":"Zhang","affiliation":""}],"abstract":"Human-AI interactivity is a critical aspect that reflects the usability of Multimodal Large Language Models (MLLMs). However, existing end-to-end MLLMs only allow users to interact with them through language instructions, leading to the limitation of the interactive accuracy and efficiency. In this study, we present precise referring instructions that utilize diverse reference representations such as points and boxes as referring prompts to refer to the special region. This enables MLLMs to focus on the region of interest and achieve finer-grained interaction. Based on precise referring instruction, we propose ChatSpot, a unified end-to-end MLLM that supports diverse forms of interactivity including mouse clicks, drag-and-drop, and drawing boxes, which provides a more flexible and seamless interactive experience. We also construct a multi-grained vision-language instruction-following dataset based on existing datasets and GPT-4 generating. Furthermore, we design a series of evaluation tasks to assess the effectiveness of region recognition and interaction. Experimental results showcase ChatSpot's promising performance. Project page: https://github.com/Ahnsun/ChatSpot.","doi":"10.24963/ijcai.2024/193","url":"https://ijcai.org/proceedings/2024/193","pdf":"https://ijcai.org/proceedings/2024/0193.pdf","markdown":"https://ijcai.org/proceedings/2024/193.md","bibtex":"https://ijcai.org/proceedings/2024/bibtex/193","volume":{"slug":"2024","title":"Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2024","url":"https://ijcai.org/proceedings/2024/","year":2024,"location":"Jeju, Korea","dates":"3-9 August 2024","editors":"Kate Larson","isbn":"978-1-956792-04-1","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2024"},"section":"Computer Vision","track":"Main Track","number":193,"pages":{"first":1743,"last":1752},"keywords":["Natural Language Processing: NLP: Question answering","Natural Language Processing: NLP: Dialogue and interactive systems","Computer Vision: CV: Embodied vision: Active agents, simulation","Computer Vision: CV: Multimodal learning"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}