Skip to yearly menu bar Skip to main content


Timezone: US/Pacific
Full Schedule Mon Tue Wed Thu Fri
Filter Events
Registration Desk
8:00 AM - 5:00 PM
Remarks
8:30 AM - 8:45 AM
Oral
8:45 AM - 10:00 AM
4 Events in this session
StreamDiffusionV2: A Streaming System for Dynamic and Interactive Video Generation
Tianrui Feng ⋅ Zhi Li ⋅ Shuo Yang ⋅ Haocheng Xi ⋅ Muyang Li ⋅ Xiuyu Li ⋅ ⋅ Keting Yang ⋅ Kelly Peng ⋅ Song Han ⋅ Maneesh Agrawala ⋅ Kurt Keutzer ⋅ Akio Kodaira ⋅ Chenfeng Xu
LEANN: A Low-Storage Overhead Vector Index
Yichuan Wang ⋅ Zhifei Li ⋅ Shu Liu ⋅ Yongji Wu ⋅ Ziming Mao ⋅ Yilong Zhao ⋅ Xiao Yan ⋅ Zhiying Xu ⋅ Yang Zhou ⋅ Ion Stoica ⋅ Sewon Min ⋅ Matei Zaharia ⋅ Joseph Gonzalez
BLASST: Dynamic BLocked Attention Sparsity via Softmax Thresholding
Jiayi Yuan ⋅ Cameron Shinn ⋅ Kai Xu ⋅ Jingze Cui ⋅ George Klimiashvili ⋅ ⋅ Perkz Zheng ⋅ Bo Li ⋅ Zhou Yuxin ⋅ Zhouhai Ye ⋅ Weijie You ⋅ Richard Cai ⋅ Julien Demouth ⋅ John D. Owens ⋅ Xia Hu ⋅ ⋅ Timmy Liu ⋅ Huizi Mao
ExecuTorch - A Unified PyTorch Solution to Run ML Models On-Device
Chen Lai ⋅ Cemal Bilgin ⋅ Abhinay Kukkadapu ⋅ Gregory Comer ⋅ ⋅ Jacob Szwejbka ⋅ ⋅ Lucy Qiu ⋅ Mengwei Liu ⋅ ⋅ Songhao Jia ⋅ ⋅ ⋅ Digant Desai ⋅ Hansong Zhang ⋅ Manuel Candales ⋅ Scott Roy ⋅ Sicheng Jia ⋅ Mergen Nachin ⋅ ⋅ Yanan Cao ⋅ Zhengxu Chen ⋅ Shunting Zhang ⋅ ⋅ Angela Yi ⋅ Zhenrui Zhang ⋅ Andrew Or ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ Supriya Rao ⋅ ⋅ Soumith Chintala
Go to Event Page
Session
8:45 AM - 10:00 AM
Invited Talk
10:30 AM - 11:30 AM

MLSys showed how systems can accelerate AI. The next shift is broader: AI is beginning to reshape the practice of systems itself. This emerging paradigm, which we call system intelligence, goes beyond automating programming tasks. It enables new forms of reasoning, design, validation, and evolution for complex systems while preserving rigor. In this talk, I will argue that system intelligence changes not only what systems we can build, but also how we understand systems as a discipline. It pushes us to rethink systems principles and methodology, shifting attention from code-level complexity to greater rigor in specification, design, and validation. Through our experiences with system verification, I will discuss how this shift may help give systems a stronger scientific foundation.

... more
Speaker Bio
Lidong Zhou
Dr. Lidong Zhou is Corporate Vice President at Microsoft, Chief Scientist of the Microsoft Asia Pacific R&D Group, and Managing Director of Microsoft Research Asia. He has held research and leadership roles across Microsoft’s Silicon Valley, Redmond, and Asia labs. His work focuses on scalable, reliable, and trustworthy distributed systems, and has award papers at SOSP, OSDI, and USENIX ATC. He has also contributed to the design of large-scale systems that power Microsoft’s search, big data, cloud, and AI infrastructure.
... more
Session
10:30 AM - 11:30 AM
Oral
1:00 PM - 2:15 PM
5 Events in this session
OSWorld-Human: Benchmarking the Efficiency of Computer-Use Agents
Reyna Abhyankar ⋅ Qi Qi ⋅ Yiying Zhang
VeriMoA: A Mixture-of-Agents Framework for Spec-to-HDL Generation
Heng Ping ⋅ Arijit Bhattacharjee ⋅ Peiyu Zhang ⋅ Shixuan Li ⋅ Wei Yang ⋅ Anzhe Cheng ⋅ Xiaole Zhang ⋅ Jesse Thomason ⋅ Ali Jannesari ⋅ Nesreen Ahmed ⋅ Paul Bogdan
When Enough is Enough: Rank-Aware Early Termination for Vector Search
Jianan Lu ⋅ Asaf Cidon ⋅ Michael None Freedman
Optimizing PyTorch Inference with LLM-Based Multi-Agent Systems
Kirill Nagaitsev ⋅ Luka Grbcic ⋅ Samuel Williams ⋅ Costin Iancu
Matrix: Peer-to-Peer Multi-Agent Synthetic Data Generation Framework
Dong Wang ⋅ Yang Li ⋅ Ansong Ni ⋅ Ching-Feng Yeh ⋅ Youssef Emad ⋅ Xinjie Lei ⋅ Liam Robbins ⋅ Karthik Padthe ⋅ Hu Xu ⋅ Xian Li ⋅ Asli Celikyilmaz ⋅ Ramya Raghavendra ⋅ LIFEI HUANG ⋅ Carole-Jean Wu ⋅ Shang-Wen Li
Go to Event Page
Oral
1:00 PM - 2:15 PM
5 Events in this session
G-HEMP: FAST MULTI-GPU PRIVATE INFERENCE FOR LARGE-SCALE GCNS WITH HOMOMORPHIC ENCRYPTION
Ran Ran ⋅ Zhaoting Gong ⋅ Zhaowei Li ⋅ Xianting Lu ⋅ Jiajia Li ⋅ Wujie Wen
db-SP: Accelerating Sparse Attention for Visual Generative Models with Dual-Balanced Sequence Parallelism
Siqi Chen ⋅ Ke Hong ⋅ Tianchen Zhao ⋅ Ruiqi Xie ⋅ Zhenhua Zhu ⋅ Xudong Zhang ⋅ Yu Wang
Flash3DGS: Algorithm and System Co-Optimization for Fast 3D Gaussian Splatting on GPUs
Lingjun Gao ⋅ Zhican Wang ⋅ Zhiwen Mo ⋅ Hongxiang Fan
TriInfer: Hybrid EPD Disaggregation for Efficient Multimodal Large Language Model Inference
Xianzhe Dong ⋅ Tongxuan Liu ⋅ Yuting Zeng ⋅ Weizhe Huang ⋅ ⋅ Siyu Wu ⋅ ⋅ Liu Yang ⋅ Yu Wu ⋅ Hailong Yang ⋅ ⋅ Jing Li
TiDAR: Think in Diffusion, Talk in Autoregression
Jingyu Liu ⋅ Xin Dong ⋅ Zhifan Ye ⋅ ⋅ Yonggan Fu ⋅ vartika singh ⋅ Ce Zhang ⋅ Pavlo Molchanov
Go to Event Page
Oral
2:30 PM - 4:00 PM
6 Events in this session
HELIOS : Adaptive Model And Early-Exit Selection for Efficient LLM Inference Serving
Avinash Kumar ⋅ Shashank Nag ⋅ Jason Clemons ⋅ LIZY JOHn ⋅ Poulami Das
Stream2LLM: Overlap Context Streaming and Prefill for Reduced Time-to-First-Token
Rajveer Bachkaniwala ⋅ ⋅ Richard So ⋅ Divya Mahajan ⋅ Kexin Rong
SuperInfer: SLO-Aware Rotary Scheduling and Memory Management for LLM Inference on Superchips
Jiahuan Yu ⋅ Mingtao Hu ⋅ ⋅ Minjia Zhang
PLA-Serve: A Prefill-Length-Aware LLM Serving System
Jianshu She ⋅ Zonghang Li ⋅ HONGCHAO DU ⋅ Shangyu Wu ⋅ Wenhao Zheng ⋅ Eric Xing ⋅ Zhengzhong Liu ⋅ Huaxiu Yao ⋅ Chun Jason Xue ⋅ Qirong Ho
TokenBlend: Accelerating Tensor Parallelism LLM Inference Through Efficient Compute-Communication Overlap
Raja Gond ⋅ Nipun Kwatra ⋅ Ramachandran Ramjee
From Tokens to Layers: Redefining Stall-Free Scheduling for LLM Serving with Layered Prefill
Gunjun Lee ⋅ ⋅ ⋅ Younjoo Lee ⋅ Jung Ho Ahn
Go to Event Page
Oral
2:30 PM - 4:00 PM
6 Events in this session
Hippocampus: An Efficient and Scalable Memory Module for Agentic AI
Yi Li ⋅ Lianjie Cao ⋅ Faraz Ahmed ⋅ Puneet Sharma ⋅ Bingzhe Li
ContextPilot: Fast Long-Context Inference via Context Reuse
Yinsicheng Jiang ⋅ Yeqi Huang ⋅ Liang Cheng ⋅ Cheng Deng ⋅ Xuan Sun ⋅ Luo Mai
AgenticCache: Cache-Driven Asynchronous Planning for Embodied AI Agents
Hojoon Kim ⋅ Yuheng Wu ⋅ Thierry Tambe
RagInfer: Efficient Retrieval-Augmented Generation Inference with Lookahead Retrieval
Chien-Yu Lin ⋅ Keisuke Kamahori ⋅ Yiyu Liu ⋅ Xiaoxiang Shi ⋅ Madhav Kashyap ⋅ Yile Gu ⋅ Rulin Shao ⋅ Zihao Ye ⋅ Kan Zhu ⋅ Rohan Kadekodi ⋅ Stephanie Wang ⋅ Arvind Krishnamurthy ⋅ Luis Ceze ⋅ Baris Kasikci
FlashAgents: Accelerating Multi-Agent LLM Systems via Streaming Prefill Overlap
Taosong Fang ⋅ Zhen Zheng ⋅ Zhengzhao Ma ⋅ Yaojie Lu ⋅ Hongyu Lin ⋅ Xianpei Han ⋅ Le Sun
Ontology-Guided Long-Term Agent Memory for Conversational RAG
Sharon Cao ⋅ Rui Li
Go to Event Page
Oral
4:30 PM - 6:00 PM
5 Events in this session
CSLE: A Reinforcement Learning Platform for Autonomous Security Management
Kim Hammar
Privatar: Scalable Privacy-preserving Multi-user VR via Secure Offloading
Jianming Tong ⋅ Hanshen Xiao ⋅ ⋅ Hao Kang ⋅ Ashish Sirasao ⋅ Ziqi Zhang ⋅ G. Edward Suh ⋅ Tushar Krishna
ZK-APEX: ZERO-KNOWLEDGE APPROXIMATE PERSONALIZED UNLEARNING WITH EXECUTABLE PROOFS
Mohammad M Maheri ⋅ ⋅ ⋅ Hamed Haddadi
Blueprint, Bootstrap, and Bridge: A Security Look at NVIDIA GPU Confidential Computing
Zhongshu Gu ⋅ ⋅ Salman Ahmed ⋅ Julian James stephen ⋅ ⋅ ⋅ Shixuan Zhao ⋅ Zhiqiang Lin
Toward Principled LLM Safety Testing: Solving the Jailbreak Oracle Problem
Shuyi Lin ⋅ Anshuman Suri ⋅ Alina Oprea ⋅ Cheng Tan
Go to Event Page
Oral
4:30 PM - 6:00 PM
6 Events in this session
SONAR: Benchmarking Topology and Collaboration in Decentralized Learning
Joyce Yuan ⋅ Yichuan Shi ⋅ Abhishek Singh ⋅ Rishi Sharma ⋅ Ramesh Raskar ⋅ ⋅ Martin Jaggi
ProToken: Token-Level Attribution for Federated Large Language Models
Waris Gill ⋅ Ahmad Humayun ⋅ Ali Anwar ⋅ Muhammad Ali Gulzar
FLoRIST: Singular Value Thresholding for Efficient and Accurate Federated Fine-Tuning of Large Language Models
Hariharan Ramesh ⋅ Jyotikrishna Dass
DisAgg: Distributed Aggregators for Efficient Secure Aggregation
Haaris Mehmood ⋅ Giorgos Tatsis ⋅ Dimitrios Alexopoulos ⋅ Karthikeyan Saravanan ⋅ Jie Xi ⋅ ⋅ Mete Ozay
PLayer-FL: A Principled Approach to Personalized Layer-wise Cross-Silo Federated Learning
Ahmed Elhussein ⋅ Florent Pollet ⋅ Gamze Gursoy
Zero redundancy distributed learning with differential privacy
Zhiqi Bu ⋅ Justin Chiu ⋅ Ruixuan Liu ⋅ Sheng Zha ⋅ George Karypis
Go to Event Page
Poster
6:00 PM - 8:00 PM
34 Events in this session
AgenticCache: Cache-Driven Asynchronous Planning for Embodied AI Agents
Hojoon Kim ⋅ Yuheng Wu ⋅ Thierry Tambe
Blueprint, Bootstrap, and Bridge: A Security Look at NVIDIA GPU Confidential Computing
Zhongshu Gu ⋅ ⋅ Salman Ahmed ⋅ Julian James stephen ⋅ ⋅ ⋅ Shixuan Zhao ⋅ Zhiqiang Lin
ContextPilot: Fast Long-Context Inference via Context Reuse
Yinsicheng Jiang ⋅ Yeqi Huang ⋅ Liang Cheng ⋅ Cheng Deng ⋅ Xuan Sun ⋅ Luo Mai
CSLE: A Reinforcement Learning Platform for Autonomous Security Management
Kim Hammar
db-SP: Accelerating Sparse Attention for Visual Generative Models with Dual-Balanced Sequence Parallelism
Siqi Chen ⋅ Ke Hong ⋅ Tianchen Zhao ⋅ Ruiqi Xie ⋅ Zhenhua Zhu ⋅ Xudong Zhang ⋅ Yu Wang
DisAgg: Distributed Aggregators for Efficient Secure Aggregation
Haaris Mehmood ⋅ Giorgos Tatsis ⋅ Dimitrios Alexopoulos ⋅ Karthikeyan Saravanan ⋅ Jie Xi ⋅ ⋅ Mete Ozay
FlashAgents: Accelerating Multi-Agent LLM Systems via Streaming Prefill Overlap
Taosong Fang ⋅ Zhen Zheng ⋅ Zhengzhao Ma ⋅ Yaojie Lu ⋅ Hongyu Lin ⋅ Xianpei Han ⋅ Le Sun
FLoRIST: Singular Value Thresholding for Efficient and Accurate Federated Fine-Tuning of Large Language Models
Hariharan Ramesh ⋅ Jyotikrishna Dass
From Tokens to Layers: Redefining Stall-Free Scheduling for MoE Serving with Layered Prefill
Gunjun Lee ⋅ ⋅ ⋅ Younjoo Lee ⋅ Jung Ho Ahn
G-HEMP: FAST MULTI-GPU PRIVATE INFERENCE FOR LARGE-SCALE GCNS WITH HOMOMORPHIC ENCRYPTION
Ran Ran ⋅ Zhaoting Gong ⋅ Zhaowei Li ⋅ Xianting Lu ⋅ Jiajia Li ⋅ Wujie Wen
HELIOS : Adaptive Model And Early-Exit Selection for Efficient LLM Inference Serving
Avinash Kumar ⋅ Shashank Nag ⋅ Jason Clemons ⋅ LIZY JOHn ⋅ Poulami Das
Hippocampus: An Efficient and Scalable Memory Module for Agentic AI
Yi Li ⋅ Lianjie Cao ⋅ Faraz Ahmed ⋅ Puneet Sharma ⋅ Bingzhe Li
Matrix: Peer-to-Peer Multi-Agent Synthetic Data Generation Framework
Dong Wang ⋅ Yang Li ⋅ Ansong Ni ⋅ Ching-Feng Yeh ⋅ Youssef Emad ⋅ Xinjie Lei ⋅ Liam Robbins ⋅ Karthik Padthe ⋅ Hu Xu ⋅ Xian Li ⋅ Asli Celikyilmaz ⋅ Ramya Raghavendra ⋅ LIFEI HUANG ⋅ Carole-Jean Wu ⋅ Shang-Wen Li
Ontology-Guided Long-Term Agent Memory for Conversational RAG
Sharon Cao ⋅ Rui Li
Optimizing PyTorch Inference with LLM-Based Multi-Agent Systems
Kirill Nagaitsev ⋅ Luka Grbcic ⋅ Samuel Williams ⋅ Costin Iancu
OSWorld-Human: Benchmarking the Efficiency of Computer-Use Agents
Reyna Abhyankar ⋅ Qi Qi ⋅ Yiying Zhang
PLA-Serve: A Prefill-Length-Aware LLM Serving System
Jianshu She ⋅ Zonghang Li ⋅ HONGCHAO DU ⋅ Shangyu Wu ⋅ Wenhao Zheng ⋅ Eric Xing ⋅ Zhengzhong Liu ⋅ Huaxiu Yao ⋅ Chun Jason Xue ⋅ Qirong Ho
PLayer-FL: A Principled Approach to Personalized Layer-wise Cross-Silo Federated Learning
Ahmed Elhussein ⋅ Florent Pollet ⋅ Gamze Gursoy
Privatar: Scalable Privacy-preserving Multi-user VR via Secure Offloading
Jianming Tong ⋅ Hanshen Xiao ⋅ ⋅ Hao Kang ⋅ Ashish Sirasao ⋅ Ziqi Zhang ⋅ G. Edward Suh ⋅ Tushar Krishna
ProToken: Token-Level Attribution for Federated Large Language Models
Waris Gill ⋅ Ahmad Humayun ⋅ Ali Anwar ⋅ Muhammad Ali Gulzar
SONAR: Benchmarking Topology and Collaboration in Decentralized Learning
Joyce Yuan ⋅ Yichuan Shi ⋅ Abhishek Singh ⋅ Rishi Sharma ⋅ Ramesh Raskar ⋅ ⋅ Martin Jaggi
Stream2LLM: Overlap Context Streaming and Prefill for Reduced Time-to-First-Token
Rajveer Bachkaniwala ⋅ ⋅ Richard So ⋅ Divya Mahajan ⋅ Kexin Rong
StreamDiffusionV2: A Streaming System for Dynamic and Interactive Video Generation
Tianrui Feng ⋅ Zhi Li ⋅ Shuo Yang ⋅ Haocheng Xi ⋅ Muyang Li ⋅ Xiuyu Li ⋅ ⋅ Keting Yang ⋅ Kelly Peng ⋅ Song Han ⋅ Maneesh Agrawala ⋅ Kurt Keutzer ⋅ Akio Kodaira ⋅ Chenfeng Xu
SuperInfer: SLO-Aware Rotary Scheduling and Memory Management for LLM Inference on Superchips
Jiahuan Yu ⋅ Mingtao Hu ⋅ ⋅ Minjia Zhang
SwiftGS: Algorithm and System Co-Optimization for Fast 3D Gaussian Splatting on GPUs
Lingjun Gao ⋅ Zhican Wang ⋅ Zhiwen Mo ⋅ Hongxiang Fan
TeleRAG: Efficient Retrieval-Augmented Generation Inference with Lookahead Retrieval
Chien-Yu Lin ⋅ Keisuke Kamahori ⋅ Yiyu Liu ⋅ Xiaoxiang Shi ⋅ Madhav Kashyap ⋅ Yile Gu ⋅ Rulin Shao ⋅ Zihao Ye ⋅ Kan Zhu ⋅ Rohan Kadekodi ⋅ Stephanie Wang ⋅ Arvind Krishnamurthy ⋅ Luis Ceze ⋅ Baris Kasikci
TiDAR: Think in Diffusion, Talk in Autoregression
Jingyu Liu ⋅ Xin Dong ⋅ Zhifan Ye ⋅ ⋅ Yonggan Fu ⋅ vartika singh ⋅ Ce Zhang ⋅ Pavlo Molchanov
TokenWeave: Efficient Compute-Communication Overlap for Distributed LLM Inference
Raja Gond ⋅ Nipun Kwatra ⋅ Ramachandran Ramjee
Toward Principled LLM Safety Testing: Solving the Jailbreak Oracle Problem
Shuyi Lin ⋅ Anshuman Suri ⋅ Alina Oprea ⋅ Cheng Tan
TriInfer: Hybrid EPD Disaggregation for Efficient Multimodal Large Language Model Inference
Xianzhe Dong ⋅ Tongxuan Liu ⋅ Yuting Zeng ⋅ Weizhe Huang ⋅ ⋅ Siyu Wu ⋅ ⋅ Liu Yang ⋅ Yu Wu ⋅ Hailong Yang ⋅ ⋅ Jing Li
VeriMoA: A Mixture-of-Agents Framework for Spec-to-HDL Generation
Heng Ping ⋅ Arijit Bhattacharjee ⋅ Peiyu Zhang ⋅ Shixuan Li ⋅ Wei Yang ⋅ Anzhe Cheng ⋅ Xiaole Zhang ⋅ Jesse Thomason ⋅ Ali Jannesari ⋅ Nesreen Ahmed ⋅ Paul Bogdan
When Enough is Enough: Rank-Aware Early Termination for Vector Search
Jianan Lu ⋅ Asaf Cidon ⋅ Michael None Freedman
Zero redundancy distributed learning with differential privacy
Zhiqi Bu ⋅ Justin Chiu ⋅ Ruixuan Liu ⋅ Sheng Zha ⋅ George Karypis
ZK-APEX: ZERO-KNOWLEDGE APPROXIMATE PERSONALIZED UNLEARNING WITH EXECUTABLE PROOFS
Mohammadmahdi Maheri ⋅ ⋅ ⋅ Hamed Haddadi
Go to Event Page