{"id":1152008,"date":"2025-10-15T21:38:59","date_gmt":"2025-10-16T04:38:59","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-academic-program&p=1152008"},"modified":"2026-01-08T00:55:02","modified_gmt":"2026-01-08T08:55:02","slug":"microsoft-research-asia-ace-talk","status":"publish","type":"msr-academic-program","link":"https:\/\/www.microsoft.com\/en-us\/research\/academic-program\/microsoft-research-asia-ace-talk\/","title":{"rendered":"Microsoft Research Asia ACE Talk"},"content":{"rendered":"\n\n

<\/p>\n\n\n\n\n\n\n

About ACE Talk<\/h4>\n\n\n\n

ACE Talk<\/strong> stands for Accelerate, Create, and Empower<\/strong> \u2014 three pillars that embody our vision to advance research and innovation across disciplines.
Launched in 2022 by Microsoft Research Asia (MSRA)<\/strong>, the ACE Talk Series provides an international platform<\/strong> for faculty, researchers, and emerging scholars to share cutting-edge ideas, spark innovation, and inspire one another toward meaningful impact.<\/p>\n\n\n\n

Mission & Vision<\/h4>\n\n\n\n

Building on this vision, the ACE Talk Series aims to:<\/p>\n\n\n\n

    \n
  • Accelerate<\/strong> the adoption of pioneering research and emerging technologies.<\/li>\n\n\n\n
  • Create<\/strong> a vibrant environment that nurtures collaboration and novel ideas.<\/li>\n\n\n\n
  • Empower<\/strong> individuals to drive innovation and positive change at scale.<\/li>\n<\/ul>\n\n\n\n

    Through this initiative, MSRA connects with rising research stars worldwide, strengthening its global academic presence while fostering a culture of openness, creativity, and cross-disciplinary exchange.<\/p>\n\n\n\n

    For MSRA researchers, ACE Talk serves as a platform to explore new frontiers, discover potential collaborators, and amplify their global influence.
    For interns and young researchers, it provides direct access to world-leading experts and invaluable career inspiration.<\/p>\n\n\n\n

    Impact & Highlights<\/h4>\n\n\n\n

    Since its inception, ACE Talk has hosted 40+ sessions<\/strong> covering a wide range of fields, including Machine Learning, Deep Learning, Robotics, Embodied AI, Visual Computing, and Social Computing<\/strong>.<\/p>\n\n\n\n

    Our speakers have represented 20+ leading universities and research institutions<\/strong> around the world \u2014 including Stanford, MIT, Princeton, CMU, UCLA, UC Berkeley, NUS, NTU, HKU, and Tsinghua University<\/strong>.<\/p>\n\n\n\n

    By the numbers:<\/strong><\/p>\n\n\n\n

      \n
    • Over 80+<\/strong> speakers approached, with 40+<\/strong> successful talks delivered.<\/li>\n\n\n\n
    • More than 10,000 total participants<\/strong>, with individual sessions drawing up to 400+ attendees<\/strong>.<\/li>\n\n\n\n
    • A growing global community connecting scholars, innovators, and practitioners across continents<\/li>\n<\/ul>\n\n\n\n

      Interested in Speaking at ACE Talk?<\/h4>\n\n\n\n

      ACE Talk welcomes faculty members, researchers, and Postdoc\/PhD students<\/strong> who are interested in sharing their research with the MSRA research community and engaging with a broader academic audience across Asia.<\/p>\n\n\n\n

      If you would like to be considered as an invited speaker, please feel free to contact us at [v-skyedu@microsoft.com] with a brief introduction and your research interests.<\/p>\n\n\n\n

      <\/div>\n\n\n\n\n\n

      Special Issues<\/h3>\n\n\n\n
      \n
      \n
      \"a<\/figure>\n\n\n\n

      Raj Reddy<\/strong><\/p>\n\n\n\n

      Turing Award Laureate<\/p>\n\n\n\n

      Talk Title: Fireside Chat with Raj Reddy, Turing Award Winner<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Yi Ma<\/strong><\/p>\n\n\n\n

      University of Hong Kong<\/p>\n\n\n\n

      Talk Title: The Past, Present, and Future of Artificial Intelligence: from black-box to white-box, from open-loop to closed-loop<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Mark D. Hill<\/strong><\/p>\n\n\n\n

      University of Wisconsin-Madison<\/p>\n\n\n\n

      Talk Title: In Computer Architecture, We Don\u2019t Change the Questions, We Change the Answers<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"Jason<\/figure>\n\n\n\n

      Jason Cong<\/strong><\/p>\n\n\n\n

      University of California, Los Angeles<\/p>\n\n\n\n

      Talk Title: Deep Learning Meets Chip Design: Driving Next-Gen Efficiency and Innovation<\/em><\/p>\n<\/div>\n<\/div>\n\n\n\n

      <\/div>\n\n\n\n

      Regular Series<\/h3>\n\n\n\n
      \n
      \n
      \"a<\/figure>\n\n\n\n

      Yuhong Zhong<\/strong><\/p>\n\n\n\n

      Columbia University<\/p>\n\n\n\n

      Talk Title: XRP: In-Kernel Storage Functions with eBPF<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Neil Gong<\/strong><\/p>\n\n\n\n

      Duke University<\/p>\n\n\n\n

      Talk Title: Big Security Issues of Big Foundation Models<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Tao Yu<\/strong><\/p>\n\n\n\n

      University of Hong Kong<\/p>\n\n\n\n

      Talk Title: Building Natural Language Interfaces through Grounding Language Models into Executable<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Jieyu Zhao<\/strong><\/p>\n\n\n\n

      University of Southern California<\/p>\n\n\n\n

      Talk Title: Building Accountable NLP Models: on Social Bias Detection and Mitigation<\/em><\/p>\n<\/div>\n<\/div>\n\n\n\n

      \n
      \n
      \"a<\/figure>\n\n\n\n

      Song Han<\/strong><\/p>\n\n\n\n

      Massachusetts Institute of Technology<\/p>\n\n\n\n

      Talk Title: SmoothQuant and AWQ for LLM quantization.<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Diyi Yang<\/strong><\/p>\n\n\n\n

      Stanford University<\/p>\n\n\n\n

      Talk Title: Challenge and Progress towards Socially Responsible NLP<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Srijan Kumar<\/strong><\/p>\n\n\n\n

      Georgia Tech<\/p>\n\n\n\n

      Talk Title: The Robustness and Reliability of Large Language and Multimodal Models<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Haiyi Zhu<\/strong><\/p>\n\n\n\n

      Carnegie Mellon University<\/p>\n\n\n\n

      Talk Title: Harmonizing Humanity and Technology: Integrating Human Values in AI-Supported Social Systems<\/em><\/p>\n<\/div>\n<\/div>\n\n\n\n

      \n
      \n
      \"a<\/figure>\n\n\n\n

      Dongkuan Xu<\/strong><\/p>\n\n\n\n

      North Carolina State University<\/p>\n\n\n\n

      Talk Title: Sculpting the Future of Collective Growth in Collaborative AI: Gentopia.AI Meets ReWOO<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Yuan-Sen Ding<\/strong><\/p>\n\n\n\n

      Australian National University & Ohio State University<\/p>\n\n\n\n

      Talk Title: From Stars to Syntax: Leveraging Deep Learning and Large Language Models for Astronomical Insights<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Hancheng Cao<\/strong><\/p>\n\n\n\n

      Stanford University<\/p>\n\n\n\n

      Talk Title: Can large language models provide useful feedback on research papers? A large-scale empirical analysis<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Weixin Liang<\/strong><\/p>\n\n\n\n

      Stanford University<\/p>\n\n\n\n

      Talk Title: Can large language models provide useful feedback on research papers? A large-scale empirical analysis<\/em><\/p>\n<\/div>\n<\/div>\n\n\n\n

      \n
      \n
      \"a<\/figure>\n\n\n\n

      Bo Li<\/strong><\/p>\n\n\n\n

      University of Chicago & University of Illinois at Urbana-Champaign<\/p>\n\n\n\n

      Talk Title: Assessing Trustworthiness and Risks of Generative Models<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      William Wang<\/strong><\/p>\n\n\n\n

      University of California, Santa Barbara<\/p>\n\n\n\n

      Talk Title: Principles of Reasoning: Compositional and Collaborative Generative AI Design<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Zsolt Kira<\/strong><\/p>\n\n\n\n

      Georgia Institute of Technology<\/p>\n\n\n\n

      Talk Title: Computer Vision in the era of Foundation Models: Progress Made and Open Challenge<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Quanquan Gu<\/strong><\/p>\n\n\n\n

      University of California, Los Angeles<\/p>\n\n\n\n

      Talk Title: Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models<\/em><\/p>\n<\/div>\n<\/div>\n\n\n\n

      \n
      \n
      \"a<\/figure>\n\n\n\n

      Max Kreminski<\/strong><\/p>\n\n\n\n

      Midjourney & Santa Clara University<\/p>\n\n\n\n

      Talk Title: Creative AI: Lensing the Imagination<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Ji Zhang<\/strong><\/p>\n\n\n\n

      Carnegie Mellon University<\/p>\n\n\n\n

      Talk Title: Autonomous Exploration and Navigation, Full Autonomy System, and Beyond<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Peng Cui<\/strong><\/p>\n\n\n\n

      Tsinghua University<\/p>\n\n\n\n

      Talk Title: Generalization and Evaluation Perspectives on the Trustworthiness of Artificial Intelligence<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Freda Shi<\/strong><\/p>\n\n\n\n

      University of Waterloo<\/p>\n\n\n\n

      Talk Title: Learning Language Structures through Grounding and Beyond<\/em><\/p>\n<\/div>\n<\/div>\n\n\n\n

      \n
      \n
      \"a<\/figure>\n\n\n\n

      Horald Soh<\/strong><\/p>\n\n\n\n

      National University of Singapore<\/p>\n\n\n\n

      Talk Title: Bridging Physical and Social Intelligence with Generative Embodied AI<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Fengli Xu<\/strong><\/p>\n\n\n\n

      Tsinghua University<\/p>\n\n\n\n

      Talk Title: Designing Efficient Reasoning and Automatic Optimization Frameworks for LLM Agents<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Beidi Chen<\/strong><\/p>\n\n\n\n

      Carnegie Mellon University<\/p>\n\n\n\n

      Talk Title: MagicPIG & Factor: Rethinking the Efficiency and Capabilities of Long-Context LLMs<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Abhishek Gupta<\/strong><\/p>\n\n\n\n

      University of Washington<\/p>\n\n\n\n

      Talk Title: Real-to-Sim-to-Real: A Scalable Technique for Robot Learning from Off-Domain Data<\/em><\/p>\n<\/div>\n<\/div>\n\n\n\n

      \n
      \n
      \"a<\/figure>\n\n\n\n

      Christina Lee Yu<\/strong><\/p>\n\n\n\n

      Cornell University<\/p>\n\n\n\n

      Talk Title: Causal Inference in the Presence of Network Interference with Low-Order Interactions<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Natasha Jaques<\/strong><\/p>\n\n\n\n

      University of Washington<\/p>\n\n\n\n

      Talk Title: Social Reinforcement Learning for pluralistic alignment and human-AI interaction<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Christopher Matthew De Sa<\/strong><\/p>\n\n\n\n

      Cornell University<\/p>\n\n\n\n

      Talk Title: Example Selection and Post-Training Quantization<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Fei Fang<\/strong><\/p>\n\n\n\n

      Carnegie Mellon University<\/p>\n\n\n\n

      Talk Title: Tackling Societal Challenges with Multi-Agent Systems: Bridging Theory and Practice<\/em><\/p>\n<\/div>\n<\/div>\n\n\n\n

      \n
      \n
      \"a<\/figure>\n\n\n\n

      Qianxiao Li<\/strong><\/p>\n\n\n\n

      National University of Singapore<\/p>\n\n\n\n

      Talk Title: Learning, approximation and control<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Hao Zhang<\/strong><\/p>\n\n\n\n

      University of California, San Diego<\/p>\n\n\n\n

      Talk Title: Fast Video Generation with Attention Tile<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Angela Yao<\/strong><\/p>\n\n\n\n

      National University of Singapore<\/p>\n\n\n\n

      Talk Title: Text, Touch and Trajectories: Reconstructing 3D Humans with 3T\u2019s<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Zongqing Lu<\/strong><\/p>\n\n\n\n

      Peking University<\/p>\n\n\n\n

      Talk Title: Scaling Humanoid Robot Learning with Internet Videos<\/em><\/p>\n<\/div>\n<\/div>\n\n\n\n

      \n
      \n
      \"a<\/figure>\n\n\n\n

      Scott Sanner<\/strong><\/p>\n\n\n\n

      University of Toronto<\/p>\n\n\n\n

      Talk Title: Verifiable, debuggable, and repairable formal reasoning with LLMs<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Amy X. Zhang<\/strong><\/p>\n\n\n\n

      University of Washington<\/p>\n\n\n\n

      Talk Title: Steering AI Agents: From Individual to Societal Scale<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Ranjay Krishna<\/strong><\/p>\n\n\n\n

      University of Washington<\/p>\n\n\n\n

      Talk Title: Prioritizing Perception in Multimodal Language Models<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Georgia Gkioxari<\/strong><\/p>\n\n\n\n

      California Institute of Technology<\/p>\n\n\n\n

      Talk Title: Scaling 3D Data to enable 3D models<\/em><\/p>\n<\/div>\n<\/div>\n\n\n\n

      \n
      \n
      \"a<\/figure>\n\n\n\n

      Yakun Sophia Shao<\/strong><\/p>\n\n\n\n

      University of California, Berkeley<\/p>\n\n\n\n

      Talk Title: From Algorithms to Silicon: Accelerating Full-Stack Co-Design for the AI Era<\/em><\/p>\n<\/div>\n\n\n\n

      \n
      \"a<\/figure>\n\n\n\n

      Kun Yuan<\/strong><\/p>\n\n\n\n

      Peking University<\/p>\n\n\n\n

      Talk Title: Memory-Efficient Training Methods for Large Language Models<\/em><\/p>\n<\/div>\n\n\n\n

      <\/div>\n\n\n\n
      <\/div>\n<\/div>\n\n\n","protected":false},"featured_media":1152314,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr_hide_image_in_river":null,"footnotes":""},"msr-opportunity-type":[155534],"msr-region":[197903],"msr-locale":[268875],"msr-program-audience":[243724],"msr-post-option":[],"msr-impact-theme":[],"class_list":["post-1152008","msr-academic-program","type-msr-academic-program","status-publish","has-post-thumbnail","hentry","msr-opportunity-type-academic-resources","msr-region-asia-pacific","msr-locale-en_us","msr-program-audience-students"],"msr_description":"","msr_social_media":[],"related-researchers":[],"tab-content":[],"msr_impact_theme":[],"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-academic-program\/1152008","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-academic-program"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-academic-program"}],"version-history":[{"count":45,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-academic-program\/1152008\/revisions"}],"predecessor-version":[{"id":1159823,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-academic-program\/1152008\/revisions\/1159823"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media\/1152314"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=1152008"}],"wp:term":[{"taxonomy":"msr-opportunity-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-opportunity-type?post=1152008"},{"taxonomy":"msr-region","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-region?post=1152008"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=1152008"},{"taxonomy":"msr-program-audience","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-program-audience?post=1152008"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=1152008"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=1152008"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}