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
Oral
8:30 AM - 10:00 AM
6 Events in this session
BEAM: Joint Resource–Power Optimization for Energy-Efficient LLM Inference under SLO contraints
Hyunjae Lee ⋅ Sangjin Choi ⋅ Seungjae Lim ⋅ Youngjin Kwon
MorphServe: Efficient and Workload-Aware LLM Serving via Runtime Quantized Layer Swapping and KV Cache Resizing
Zhaoyuan Su ⋅ Zeyu Zhang ⋅ Tingfeng Lan ⋅ Zirui Wang ⋅ Haiying Shen ⋅ Juncheng Yang ⋅ Yue Cheng
BOute: Cost-Efficient LLM Serving with Heterogeneous LLMs and GPUs via Multi-Objective Bayesian Optimization
YOUHE JIANG ⋅ Fangcheng Fu ⋅ Eiko Yoneki
Locality-Aware Beam Scheduling for Efficient Test-Time Compute with a Consumer-grade GPU
Hsing-Ti Wang ⋅ Hung-Tso Shiao ⋅ Chia-Lin Yang
Breaking the Ice: Analyzing Cold Start Latency in vLLM
Huzaifa Shaaban Kabakibo ⋅ Animesh Trivedi ⋅ Lin Wang
FaaScale: Unlocking Fast LLM Scaling for Serverless Inference
Minchen Yu ⋅ Rui Yang ⋅ Chaobo Jia ⋅ Zhaoyuan Su ⋅ Sheng Yao ⋅ Tingfeng Lan ⋅ Yuchen Yang ⋅ Zirui Wang ⋅ Yue Cheng ⋅ Wei Wang ⋅ Ao Wang ⋅ Ruichuan Chen
Go to Event Page
Oral
8:30 AM - 10:00 AM
6 Events in this session
BOOST: BOttleneck-Optimized Scalable Training Framework for Low-Rank Large Language Models
Zhengyang Wang ⋅ Ziyue Liu ⋅ Ruijie Zhang ⋅ Avinash Maurya ⋅ Bogdan Nicolae ⋅ Paul Hovland ⋅ Franck Cappello ⋅ zheng Zhang
Unleashing Scalable Context Parallelism for Foundation Models Pre-Training via FCP
Yilong Zhao ⋅ Xiaonan Nie ⋅ Kan Zhu ⋅ Shuang Ma ⋅ Zhichao Lai ⋅ Hongxiang Hao ⋅ Yang Zhou ⋅ Baris Kasikci ⋅ Ion Stoica
ProTrain: Efficient LLM Training via Automatic Memory Management
Hanmei Yang ⋅ Jin Zhou ⋅ Yao Fu ⋅ Xiaoqun Wang ⋅ Ramine Roane ⋅ Hui Guan ⋅ Tongping Liu
MTraining: Distributed Dynamic Sparse Attention for Efficient Ultra-Long Context Training
Wenxuan Li ⋅ Chengruidong Zhang ⋅ Huiqiang Jiang ⋅ Yucheng Li ⋅ Yuqing Yang ⋅ Lili Qiu
NEST: Network- and Memory-Aware Device Placement for Distributed Deep Learning
Irene Wang ⋅ Vishnu Venkata ⋅ Arvind Krishnamurthy ⋅ Divya Mahajan
Efficient Long-Context Language Model Training by Core Attention Disaggregation
Yonghao Zhuang ⋅ Junda Chen ⋅ Bo Pang ⋅ Yi Gu ⋅ Yibo Zhu ⋅ Yimin Jiang ⋅ Ion Stoica ⋅ Hao Zhang ⋅ Eric Xing
Go to Event Page
Invited Talk
10:30 AM - 11:30 AM
Speaker Bio
Amin Vahdat
Amin Vahdat is the SVP and Chief Technologist, AI & Infrastructure at Google. His team is responsible for delivering industry-leading infrastructure which spans custom silicon, data centers, network, and supply chain and operations. This infrastructure serves Alphabet, Google and the world, and Artificial Intelligence technologies that empower ML developers and solve customers’ most pressing business challenges. In the past, he was Vice President and General Manager for Google's compute, storage, and network hardware and software infrastructure. Until 2019, he was the Technical Lead and Vice President for the Networking organization at Google.
... more
Session
10:30 AM - 11:30 AM
Panel

Fireside Chat

Amin Vahdat ⋅ Zhihao Jia ⋅ Aakanksha Chowdhery
11:30 AM - 12:15 PM
Oral
1:30 PM - 3:00 PM
6 Events in this session
SkipKV: Selective Skipping of KV Generation and Storage for Efficient Inference with Large Reasoning Models
Jiayi Tian ⋅ Seyedarmin Azizi ⋅ Yequan Zhao ⋅ Erfan Potraghloo ⋅ Sean McPherson ⋅ Sharath Nittur Sridhar ⋅ Zhengyang Wang ⋅ zheng Zhang ⋅ Massoud Pedram ⋅ Souvik Kundu
OPKV: A High-Throughput Plugin-Driven Framework for Recallable Sparsity in Paged KV Cache Systems
Huazheng Lao ⋅ Xiaofeng Li ⋅ Rui Xu ⋅ Long Chen ⋅ Xia Zhu ⋅ Jinquan Zhang
FlexiCache: Leveraging Temporal Stability of Attention Heads for Efficient KV Cache Management
Nazmul Takbir ⋅ Hamidreza Alikhani Koshkak ⋅ Nikil Dutt ⋅ Sangeetha Abdu Jyothi
MAC-Attention: a Match--Amend--Complete scheme for fast and accurate attention computation
Jinghan Yao ⋅ Sam Jacobs ⋅ Walid Krichene ⋅ Masahiro Tanaka ⋅ Dhabaleswar Panda
Using Span Queries to Optimize Cache and Attention Locality
Paul Castro ⋅ Nick Mitchell ⋅ Nathan Ordonez ⋅ Thomas Parnell ⋅ Mudhakar Srivatsa ⋅ Antoni Viros i Martin
Kitty: Accurate and Efficient 2-bit KV Cache Quantization with Dynamic Channel-wise Precision Boost
Haojun Xia ⋅ Xiaoxia Wu ⋅ Jisen Li ⋅ Tsai-chuan Wu ⋅ Junxiong Wang ⋅ Jue Wang ⋅ Chenxi Li ⋅ Aman Singhal ⋅ Alay Dilipbhai Shah ⋅ Alpay Ariyak ⋅ Donglin Zhuang ⋅ Zhongzhu Zhou ⋅ Ben Athiwaratkun ⋅ Zhen Zheng ⋅ Shuaiwen Song
Go to Event Page
Oral
1:30 PM - 3:00 PM
6 Events in this session
FlashInfer-Bench: Building the Virtuous Cycle for AI-driven LLM Systems
Shanli Xing ⋅ Vivian Zhai ⋅ Alexander Jiang ⋅ Yixin Dong ⋅ Yong Wu ⋅ Zihao Ye ⋅ Charlie Ruan ⋅ Yingyi Huang ⋅ Yineng Zhang ⋅ Liangsheng Yin ⋅ Aksara Bayyapu ⋅ Luis Ceze ⋅ Tianqi Chen
Virtual Machine NUMA Placement at Scale: Learning the Norm, Shielding the Tail
Yibo Zhao ⋅ Tianyuan Wu ⋅ HUI XUE ⋅ Qi Chen ⋅ Zhenhua Han ⋅ Zikai Xu ⋅ Yuntai Chang ⋅ Rui Gao ⋅ Steve Deng ⋅ Jui-Hao Chiang ⋅ Mingxia Li ⋅ Yuqing Yang ⋅ Cheng Tan ⋅ Fan Yang ⋅ Peng Cheng ⋅ Yongqiang Xiong ⋅ Lili Qiu ⋅ Lidong Zhou
When Machine Learning Isn’t Sure: Building Resilient ML-Based Computer Systems by Embracing Uncertainty
Varun Gohil ⋅ Nevena Stojkovic ⋅ Noman Bashir ⋅ Sundar Dev ⋅ Gaurang Upasani ⋅ David Lo ⋅ Parthasarathy Ranganathan ⋅ Christina Delimitrou
Practical Adversarial Multi-Armed Bandits with Sublinear Runtime
Kasper Overgaard Mortensen ⋅ Ama Bembua Bainson ⋅ Mathias Tversted ⋅ Kristoffer Strube ⋅ Renata Borovica-Gajic ⋅ Andrea Paudice ⋅ Davide Mottin ⋅ Panagiotis Karras
Unified LLM Model for Power, Performance, and Area Prediction from Hardware Code
Armin Abdollahi ⋅ Mehdi Kamal ⋅ Massoud Pedram
Automated Algorithm Design for Auto-Tuning Optimizers
Floris-Jan Willemsen ⋅ Niki van Stein ⋅ Ben van Werkhoven
Go to Event Page
Oral
3:15 PM - 4:45 PM
6 Events in this session
RaidServe: High-performance Resilient Serving
Ziyi Xu ⋅ Zhiqiang Xie ⋅ Swapnil Gandhi ⋅ Christos Kozyrakis
fabric-lib: RDMA Point-to-Point Communication for LLM Systems
Nandor Licker ⋅ Kevin Hu ⋅ Vladimir Zaytsev ⋅ Lequn Chen
Demystifying the Mixture of Experts Serving Tax
Pratyush Patel ⋅ Dayeol Lee ⋅ Shintaro Iwasaki ⋅ Arvind Krishnamurthy
FarSkip-Collective: Unhobbling Blocking Communication in Mixture of Experts Models
Yonatan Dukler ⋅ Guihong Li ⋅ Deval Shah ⋅ Jiang Liu ⋅ Vikram Appia ⋅ Emad Barsoum
GhostServe: A Lightweight Checkpointing System in the Shadow for Fault-Tolerant LLM Serving
Shakya Jayakody ⋅ Youpeng Zhao ⋅ Chinmay Dhanraj Nehate ⋅ Jun Wang
CRAFT: Fine-Grained Cost-Aware Expert Replication For Efficient Mixture-of-Experts Serving
Adrian Zhao ⋅ Zhenkun Cai ⋅ Zhenyu Song ⋅ Lingfan Yu ⋅ Haozheng Fan ⋅ Jun Wu ⋅ Yida Wang ⋅ Nandita Vijaykumar
Go to Event Page
Oral
3:15 PM - 4:45 PM
6 Events in this session
ReSpec: Towards Optimizing Speculative Decoding in Reinforcement Learning Systems
Qiaoling Chen ⋅ Zijun Liu ⋅ Peng Sun ⋅ Shenggui Li ⋅ Guoteng Wang ⋅ Ziming Liu ⋅ Yonggang Wen ⋅ Siyuan Feng ⋅ Tianwei Zhang
HetRL: Efficient Reinforcement Learning for LLMs in Heterogeneous Environments
Yongjun He ⋅ Shuai Zhang ⋅ Jiading Gai ⋅ Xiyuan Zhang ⋅ Boran Han ⋅ Bernie Wang ⋅ Huzefa Rangwala ⋅ George Karypis
Pylo: Towards Accessible Learned Optimizers in PyTorch
Paul Janson ⋅ Benjamin Thérien ⋅ Quentin Anthony ⋅ Xiaolong Huang ⋅ Abhinav Moudgil ⋅ Eugene Belilovsky
Learning from Less: Measuring the Effectiveness of RLVR in Low Data and Compute Regimes
Justin Bauer ⋅ Thomas Walshe ⋅ Derek Pham ⋅ Harit Vishwakarma ⋅ Armin Parchami ⋅ Frederic Sala ⋅ Paroma Varma
FlexTrain: Scalable Hybrid-Parallel Training with Elastic Resource Utilization and Consistent Accuracy
Weilin Cai ⋅ Diandian Gu ⋅ Baoquan Zhong ⋅ Jun Wang ⋅ Zhuolin Zheng ⋅ Gaohong Liu ⋅ Jiang Kaihua ⋅ Shuguang Wang ⋅ Wencong Xiao ⋅ Jiayi Huang
Beat the long tail: Distribution-Aware Speculative Decoding for RL Training
Zelei Shao ⋅ Vikranth Srivatsa ⋅ Sanjana Srivastava ⋅ Qingyang Wu ⋅ Alpay Ariyak ⋅ Xiaoxia Wu ⋅ Ameen Patel ⋅ Jue Wang ⋅ Percy Liang ⋅ Tri Dao ⋅ Ce Zhang ⋅ Yiying Zhang ⋅ Ben Athiwaratkun ⋅ Chenfeng Xu ⋅ Junxiong Wang
Go to Event Page
Oral
5:00 PM - 6:30 PM
6 Events in this session
Zorse: Optimizing LLM Training Efficiency on Heterogeneous GPU Clusters
Runsheng Guo ⋅ Utkarsh Anand ⋅ Khuzaima Daudjee ⋅ Rathijit Sen
GriNNder: Breaking the Memory Capacity Wall in Full-Graph GNN Training with Storage Offloading
Jaeyong Song ⋅ Seongyeon Park ⋅ Hongsun Jang ⋅ Jaewon Jung ⋅ Hunseong Lim ⋅ Junguk Hong ⋅ Jinho Lee
HexiScale: Facilitating Large Language Model Training over Heterogeneous Hardware
Ran Yan ⋅ YOUHE JIANG ⋅ Xiaonan Nie ⋅ Fangcheng Fu ⋅ Bin CUI ⋅ Binhang Yuan
A Lightweight High-Throughput Collective-Capable NoC for Large-Scale ML Accelerators
Luca Colagrande ⋅ Lorenzo Leone ⋅ Chen Wu ⋅ Tim Fischer ⋅ Raphael Roth ⋅ Luca Benini
DreamDDP: Accelerating Low-Bandwidth Geo-Distributed LLM Training with Layer-wise Partial Synchronization
Zhenheng Tang ⋅ Zichen TANG ⋅ Junlin Huang ⋅ Xinglin Pan ⋅ Rudan Yan ⋅ Yuxin Wang ⋅ Amelie Zhou ⋅ Shaohuai Shi ⋅ Xiaowen Chu ⋅ Bo Li
FP8-Flow-MoE: A Casting-Free FP8 Recipe without Double Quantization Error
Fengjuan Wang ⋅ Zhiyi Su ⋅ Xingzhu Hu ⋅ Cheng Wang ⋅ Sun Mou
Go to Event Page
Oral
5:00 PM - 6:30 PM
5 Events in this session
CAGE: Curvature-Aware Gradient Estimation For Accurate Quantization-Aware Training
Soroush Tabesh ⋅ Mher Safaryan ⋅ Andrei Panferov ⋅ Alexandra Volkova ⋅ Dan Alistarh
Shannonic: Efficient Entropy-Optimal Compression for ML Workloads
Kareem Ibrahim ⋅ Mohammadjavad Maheronnaghsh ⋅ Andreas Moshovos
Once-for-All Channel Mixers (HyperTinyPW): Generative Compression for TinyML
Yassien Shaalan
MixLLM: LLM Quantization with Global Mixed-precision between Output-features and Highly-efficient System Design
Zhen Zheng ⋅ Xiaonan Song ⋅ Chuanjie Liu
Search Your Block Floating Point Scales!
Tanmaey Gupta ⋅ Hayden Prairie ⋅ Xiaoxia Wu ⋅ Reyna Abhyankar ⋅ Qingyang Wu ⋅ Austin Silveria ⋅ Pragaash Ponnusamy ⋅ Jue Wang ⋅ Ben Athiwaratkun ⋅ Shuaiwen Song ⋅ Tri Dao ⋅ Daniel Fu ⋅ Christopher De Sa
Go to Event Page
Poster
6:30 PM - 8:30 PM
47 Events in this session
BEAM: Joint Resource–Power Optimization for Energy-Efficient LLM Inference under SLO contraints
Hyunjae Lee ⋅ Sangjin Choi ⋅ Seungjae Lim ⋅ Youngjin Kwon
BOOST: BOttleneck-Optimized Scalable Training Framework for Low-Rank Large Language Models
Zhengyang Wang ⋅ Ziyue Liu ⋅ Ruijie Zhang ⋅ Avinash Maurya ⋅ Bogdan Nicolae ⋅ Paul Hovland ⋅ Franck Cappello ⋅ zheng Zhang
Unleashing Scalable Context Parallelism for Foundation Models Pre-Training via FCP
Yilong Zhao ⋅ Xiaonan Nie ⋅ Kan Zhu ⋅ Shuang Ma ⋅ Zhichao Lai ⋅ Hongxiang Hao ⋅ Yang Zhou ⋅ Baris Kasikci ⋅ Ion Stoica
MorphServe: Efficient and Workload-Aware LLM Serving via Runtime Quantized Layer Swapping and KV Cache Resizing
Zhaoyuan Su ⋅ Zeyu Zhang ⋅ Tingfeng Lan ⋅ Zirui Wang ⋅ Haiying Shen ⋅ Juncheng Yang ⋅ Yue Cheng
ProTrain: Efficient LLM Training via Automatic Memory Management
Hanmei Yang ⋅ Jin Zhou ⋅ Yao Fu ⋅ Xiaoqun Wang ⋅ Ramine Roane ⋅ Hui Guan ⋅ Tongping Liu
BOute: Cost-Efficient LLM Serving with Heterogeneous LLMs and GPUs via Multi-Objective Bayesian Optimization
YOUHE JIANG ⋅ Fangcheng Fu ⋅ Eiko Yoneki
Locality-Aware Beam Scheduling for Efficient Test-Time Compute with a Consumer-grade GPU
Hsing-Ti Wang ⋅ Hung-Tso Shiao ⋅ Chia-Lin Yang
MTraining: Distributed Dynamic Sparse Attention for Efficient Ultra-Long Context Training
Wenxuan Li ⋅ Chengruidong Zhang ⋅ Huiqiang Jiang ⋅ Yucheng Li ⋅ Yuqing Yang ⋅ Lili Qiu
Breaking the Ice: Analyzing Cold Start Latency in vLLM
Huzaifa Shaaban Kabakibo ⋅ Animesh Trivedi ⋅ Lin Wang
NEST: Network- and Memory-Aware Device Placement for Distributed Deep Learning
Irene Wang ⋅ Vishnu Venkata ⋅ Arvind Krishnamurthy ⋅ Divya Mahajan
FaaScale: Unlocking Fast LLM Scaling for Serverless Inference
Minchen Yu ⋅ Rui Yang ⋅ Chaobo Jia ⋅ Zhaoyuan Su ⋅ Sheng Yao ⋅ Tingfeng Lan ⋅ Yuchen Yang ⋅ Zirui Wang ⋅ Yue Cheng ⋅ Wei Wang ⋅ Ao Wang ⋅ Ruichuan Chen
Efficient Long-Context Language Model Training by Core Attention Disaggregation
Yonghao Zhuang ⋅ Junda Chen ⋅ Bo Pang ⋅ Yi Gu ⋅ Yibo Zhu ⋅ Yimin Jiang ⋅ Ion Stoica ⋅ Hao Zhang ⋅ Eric Xing
SkipKV: Selective Skipping of KV Generation and Storage for Efficient Inference with Large Reasoning Models
Jiayi Tian ⋅ Seyedarmin Azizi ⋅ Yequan Zhao ⋅ Erfan Potraghloo ⋅ Sean McPherson ⋅ Sharath Nittur Sridhar ⋅ Zhengyang Wang ⋅ zheng Zhang ⋅ Massoud Pedram ⋅ Souvik Kundu
FlashInfer-Bench: Building the Virtuous Cycle for AI-driven LLM Systems
Shanli Xing ⋅ Vivian Zhai ⋅ Alexander Jiang ⋅ Yixin Dong ⋅ Yong Wu ⋅ Zihao Ye ⋅ Charlie Ruan ⋅ Yingyi Huang ⋅ Yineng Zhang ⋅ Liangsheng Yin ⋅ Aksara Bayyapu ⋅ Luis Ceze ⋅ Tianqi Chen
OPKV: A High-Throughput Plugin-Driven Framework for Recallable Sparsity in Paged KV Cache Systems
Huazheng Lao ⋅ Xiaofeng Li ⋅ Rui Xu ⋅ Long Chen ⋅ Xia Zhu ⋅ Jinquan Zhang
Virtual Machine NUMA Placement at Scale: Learning the Norm, Shielding the Tail
Yibo Zhao ⋅ Tianyuan Wu ⋅ HUI XUE ⋅ Qi Chen ⋅ Zhenhua Han ⋅ Zikai Xu ⋅ Yuntai Chang ⋅ Rui Gao ⋅ Steve Deng ⋅ Jui-Hao Chiang ⋅ Mingxia Li ⋅ Yuqing Yang ⋅ Cheng Tan ⋅ Fan Yang ⋅ Peng Cheng ⋅ Yongqiang Xiong ⋅ Lili Qiu ⋅ Lidong Zhou
FlexiCache: Leveraging Temporal Stability of Attention Heads for Efficient KV Cache Management
Nazmul Takbir ⋅ Hamidreza Alikhani Koshkak ⋅ Nikil Dutt ⋅ Sangeetha Abdu Jyothi
When Machine Learning Isn’t Sure: Building Resilient ML-Based Computer Systems by Embracing Uncertainty
Varun Gohil ⋅ Nevena Stojkovic ⋅ Noman Bashir ⋅ Sundar Dev ⋅ Gaurang Upasani ⋅ David Lo ⋅ Parthasarathy Ranganathan ⋅ Christina Delimitrou
MAC-Attention: a Match--Amend--Complete scheme for fast and accurate attention computation
Jinghan Yao ⋅ Sam Jacobs ⋅ Walid Krichene ⋅ Masahiro Tanaka ⋅ Dhabaleswar Panda
Practical Adversarial Multi-Armed Bandits with Sublinear Runtime
Kasper Overgaard Mortensen ⋅ Ama Bembua Bainson ⋅ Mathias Tversted ⋅ Kristoffer Strube ⋅ Renata Borovica-Gajic ⋅ Andrea Paudice ⋅ Davide Mottin ⋅ Panagiotis Karras
Unified LLM Model for Power, Performance, and Area Prediction from Hardware Code
Armin Abdollahi ⋅ Mehdi Kamal ⋅ Massoud Pedram
Using Span Queries to Optimize Cache and Attention Locality
Paul Castro ⋅ Nick Mitchell ⋅ Nathan Ordonez ⋅ Thomas Parnell ⋅ Mudhakar Srivatsa ⋅ Antoni Viros i Martin
Automated Algorithm Design for Auto-Tuning Optimizers
Floris-Jan Willemsen ⋅ Niki van Stein ⋅ Ben van Werkhoven
Kitty: Accurate and Efficient 2-bit KV Cache Quantization with Dynamic Channel-wise Precision Boost
Haojun Xia ⋅ Xiaoxia Wu ⋅ Jisen Li ⋅ Tsai-chuan Wu ⋅ Junxiong Wang ⋅ Jue Wang ⋅ Chenxi Li ⋅ Aman Singhal ⋅ Alay Dilipbhai Shah ⋅ Alpay Ariyak ⋅ Donglin Zhuang ⋅ Zhongzhu Zhou ⋅ Ben Athiwaratkun ⋅ Zhen Zheng ⋅ Shuaiwen Song
RaidServe: High-performance Resilient Serving
Ziyi Xu ⋅ Zhiqiang Xie ⋅ Swapnil Gandhi ⋅ Christos Kozyrakis
ReSpec: Towards Optimizing Speculative Decoding in Reinforcement Learning Systems
Qiaoling Chen ⋅ Zijun Liu ⋅ Peng Sun ⋅ Shenggui Li ⋅ Guoteng Wang ⋅ Ziming Liu ⋅ Yonggang Wen ⋅ Siyuan Feng ⋅ Tianwei Zhang
HetRL: Efficient Reinforcement Learning for LLMs in Heterogeneous Environments
Yongjun He ⋅ Shuai Zhang ⋅ Jiading Gai ⋅ Xiyuan Zhang ⋅ Boran Han ⋅ Bernie Wang ⋅ Huzefa Rangwala ⋅ George Karypis
fabric-lib: RDMA Point-to-Point Communication for LLM Systems
Nandor Licker ⋅ Kevin Hu ⋅ Vladimir Zaytsev ⋅ Lequn Chen
Pylo: Towards Accessible Learned Optimizers in PyTorch
Paul Janson ⋅ Benjamin Thérien ⋅ Quentin Anthony ⋅ Xiaolong Huang ⋅ Abhinav Moudgil ⋅ Eugene Belilovsky
Demystifying the Mixture of Experts Serving Tax
Pratyush Patel ⋅ Dayeol Lee ⋅ Shintaro Iwasaki ⋅ Arvind Krishnamurthy
Learning from Less: Measuring the Effectiveness of RLVR in Low Data and Compute Regimes
Justin Bauer ⋅ Thomas Walshe ⋅ Derek Pham ⋅ Harit Vishwakarma ⋅ Armin Parchami ⋅ Frederic Sala ⋅ Paroma Varma
FarSkip-Collective: Unhobbling Blocking Communication in Mixture of Experts Models
Yonatan Dukler ⋅ Guihong Li ⋅ Deval Shah ⋅ Jiang Liu ⋅ Vikram Appia ⋅ Emad Barsoum
FlexTrain: Scalable Hybrid-Parallel Training with Elastic Resource Utilization and Consistent Accuracy
Weilin Cai ⋅ Diandian Gu ⋅ Baoquan Zhong ⋅ Jun Wang ⋅ Zhuolin Zheng ⋅ Gaohong Liu ⋅ Jiang Kaihua ⋅ Shuguang Wang ⋅ Wencong Xiao ⋅ Jiayi Huang
GhostServe: A Lightweight Checkpointing System in the Shadow for Fault-Tolerant LLM Serving
Shakya Jayakody ⋅ Youpeng Zhao ⋅ Chinmay Dhanraj Nehate ⋅ Jun Wang
Beat the long tail: Distribution-Aware Speculative Decoding for RL Training
Zelei Shao ⋅ Vikranth Srivatsa ⋅ Sanjana Srivastava ⋅ Qingyang Wu ⋅ Alpay Ariyak ⋅ Xiaoxia Wu ⋅ Ameen Patel ⋅ Jue Wang ⋅ Percy Liang ⋅ Tri Dao ⋅ Ce Zhang ⋅ Yiying Zhang ⋅ Ben Athiwaratkun ⋅ Chenfeng Xu ⋅ Junxiong Wang
CRAFT: Fine-Grained Cost-Aware Expert Replication For Efficient Mixture-of-Experts Serving
Adrian Zhao ⋅ Zhenkun Cai ⋅ Zhenyu Song ⋅ Lingfan Yu ⋅ Haozheng Fan ⋅ Jun Wu ⋅ Yida Wang ⋅ Nandita Vijaykumar
Zorse: Optimizing LLM Training Efficiency on Heterogeneous GPU Clusters
Runsheng Guo ⋅ Utkarsh Anand ⋅ Khuzaima Daudjee ⋅ Rathijit Sen
CAGE: Curvature-Aware Gradient Estimation For Accurate Quantization-Aware Training
Soroush Tabesh ⋅ Mher Safaryan ⋅ Andrei Panferov ⋅ Alexandra Volkova ⋅ Dan Alistarh
GriNNder: Breaking the Memory Capacity Wall in Full-Graph GNN Training with Storage Offloading
Jaeyong Song ⋅ Seongyeon Park ⋅ Hongsun Jang ⋅ Jaewon Jung ⋅ Hunseong Lim ⋅ Junguk Hong ⋅ Jinho Lee
Shannonic: Efficient Entropy-Optimal Compression for ML Workloads
Kareem Ibrahim ⋅ Mohammadjavad Maheronnaghsh ⋅ Andreas Moshovos
HexiScale: Facilitating Large Language Model Training over Heterogeneous Hardware
Ran Yan ⋅ YOUHE JIANG ⋅ Xiaonan Nie ⋅ Fangcheng Fu ⋅ Bin CUI ⋅ Binhang Yuan
Once-for-All Channel Mixers (HyperTinyPW): Generative Compression for TinyML
Yassien Shaalan
MixLLM: LLM Quantization with Global Mixed-precision between Output-features and Highly-efficient System Design
Zhen Zheng ⋅ Xiaonan Song ⋅ Chuanjie Liu
A Lightweight High-Throughput Collective-Capable NoC for Large-Scale ML Accelerators
Luca Colagrande ⋅ Lorenzo Leone ⋅ Chen Wu ⋅ Tim Fischer ⋅ Raphael Roth ⋅ Luca Benini
DreamDDP: Accelerating Low-Bandwidth Geo-Distributed LLM Training with Layer-wise Partial Synchronization
Zhenheng Tang ⋅ Zichen TANG ⋅ Junlin Huang ⋅ Xinglin Pan ⋅ Rudan Yan ⋅ Yuxin Wang ⋅ Amelie Zhou ⋅ Shaohuai Shi ⋅ Xiaowen Chu ⋅ Bo Li
Search Your Block Floating Point Scales!
Tanmaey Gupta ⋅ Hayden Prairie ⋅ Xiaoxia Wu ⋅ Reyna Abhyankar ⋅ Qingyang Wu ⋅ Austin Silveria ⋅ Pragaash Ponnusamy ⋅ Jue Wang ⋅ Ben Athiwaratkun ⋅ Shuaiwen Song ⋅ Tri Dao ⋅ Daniel Fu ⋅ Christopher De Sa
FP8-Flow-MoE: A Casting-Free FP8 Recipe without Double Quantization Error
Fengjuan Wang ⋅ Zhiyi Su ⋅ Xingzhu Hu ⋅ Cheng Wang ⋅ Sun Mou
Go to Event Page