最新动态-【重磅】Tat-Seng Chua教授将在ChinaMM 2024作特邀报告！！！

学术动态

【重磅】Tat-Seng Chua教授将在ChinaMM 2024作特邀报告！！！

CCF多媒体专委会 2024年06月27日 21:58 北京

2024中国多媒体大会（ChinaMM 2024）将于2024年7月24日-26日在宁夏回族自治区银川市召开。现已开放报名注册。

大会邀请Tat-Seng Chua教授作特邀报告，期待与大家共享学术盛宴！

嘉宾简介

Dr. Chua is the KITHCT Chair Professor at the School of Computing, National University of Singapore (NUS). He is also the Distinguished Visiting Professor of Tsinghua University, the Visiting Pao Yue-Kong Chair Professor of Zhejiang University, and the Distinguished Visiting Professor of Sichuan University. Dr. Chua was the Founding Dean of the School of Computing from 1998-2000. His main research interests include unstructured data analytics, video analytics, conversational search and recommendation, and robust and trustable AI. He is the co-Director of NExT, a joint research Center between NUS and Tsinghua University.

Dr. Chua is the recipient of the 2015 ACM SIGMM Achievements Award, and the winner of the 2022 NUS Research Recognition Award. He is the Chair of steering committee of Multimedia Modeling (MMM) conference series, and ACM International Conference on Multimedia Retrieval (ICMR) (2015-2018). He was the General Co-Chair of ACM Multimedia 2005, ACM SIGIR 2008, ACM Web Science 2015, ACM MM-Asia 2020, WSDM 2023, and TheWebConf (or WWW) 2024. He serves in the editorial boards of several international journals. Dr. Chua is the co-Founder of two technology startup companies in Singapore.

报告简介

No.1

报告题目

From Large Foundation Models to Network of Experts

No.2

报告摘要

The emergence of Large Foundation Models (LFM’s) that offer significant capabilities in content comprehension, generation, and flexible human-level dialogues, has revolutionized the way we seek and consume information. At sufficiently large scale, universal LFMs possess the strengths of domain versatility, output diversity, as well as semantic coherency and alignment at human-level. But being too versatile, diverse, large and open-domain come at the price of requiring high cost for operation and hallucination with the accompanying problems of trust and safety. There is a need to develop efficient solutions that can retain most of the strengths of LFMs while being safe, trustable and accurate in solving task-specific problems. This talk presents recent research and trends of LFMs at two levels. The first is at the media level, by analyzing recent advancements in aligning text and multimodal contents to achieve higher quality multimodal understanding and generation, as well as research towards multimodal alignment, instruction tuning and reliable & faithful LFMs. At the next level, we discuss current trends towards agent-based systems, and present our proposal to advance the agent-based models towards the network of experts (NoE) framework. The NoE will be task specific, for tasks such as recommendation, event forecasting and Fintech etc., while domain agnostic. It will support a network of small-sized LFMs or experts with diverse capabilities; some with high performance and with high level of trust and safety on domain specific tasks; while others are domain agnostic and with high versatility and diversity. These experts can be configured dynamically to collaboratively tackle more complex problems beyond what each is capable of doing. As NoE requires only smaller sized LFMs tuned to solving various tasks, they can be undertaken by smaller academic and industry labs, and thus offering the possibility of democratizing the core of LFM research to all.

会议简介

由中国图象图形学学会（CSIG）、中国计算机学会（CCF）主办，CSIG多媒体专业委员会、CCF多媒体技术专业委员会承办，北方民族大学联合承办，宁夏大学、宁夏师范学院协办的中国多媒体大会（ChinaMM 2024），将于2024年7月24-26日在宁夏回族自治区银川市召开。

本次大会将为多媒体领域的学术界和产业界提供前沿的技术发展动态、丰富多样的学术交流活动、全面前瞻的产业技术宣传平台，以广泛促进学术分享与交流、产业合作与互动，推动产学研联动发展，提升我国多媒体技术的研究与应用水平。ChinaMM 2024期待和更多企业展开深度合作，携手共同促进多媒体技术研究与产业发展。

会议注册

注册方式

No.1

电脑端

电脑端请通过以下链接进行注册：https://conf.csig.org.cn/10395.html

No.2

手机端

手机端请通过以下二维码进行注册：

No.3

注册费用

（1）如果您有多篇论文，需要用学生会员身份注册录用文章，请选择“学生会员（论文注册）”

（2）“学生注册”为全日制在校学生，不含非全日制研究生和博士后，注册须提供学生身份证明

（3）CSIG / CCF 会员（含学生会员）注册必须使用会员系统预留手机号码，请准确填写手机号码和姓名

（4）无特殊情况，每篇录用论文至少1位作者必须注册非学生类型注册。同一作者如有多篇论文录用，需要独立注册

（5）加入CSIG会员请点击 https://www.csig.org.cn/30/201807/49383.html（申请入会时请选择“多媒体专委会”），加入CCF会员请点击https://www.ccf.org.cn/Membership/Individual_member/How_to_Join/ （申请入会时请选择“多媒体技术专委会”）。对于非会员人员，建议先申请加入学会，获得会员号后享受优惠价格注册会议。