Skip to yearly menu bar Skip to main content


Show Detail
Schedule Mon Tue Wed Thu
Timezone: US/Pacific
 
Filter Rooms:  

SUN 11 MAY
3 p.m.
Registration and Check-in
(ends 5:00 PM)

MON 12 MAY
8:30 a.m.
Registration and coffee
(ends 4:00 PM)
9:20 a.m.
Opening Remarks - Young Professional Symposium
(ends 9:30 AM)
9:30 a.m.
Invited Talk:
Extreme PyTorch: Inside the Most Demanding ML Workloads—and the Open Challenges in Building AI Agents to Democratize Them
Soumith Chintala
(ends 10:30 AM)
10:30 a.m.
Break:
Break
(ends 10:45 AM)
10:45 a.m.
Talk:
Lessons Learned from Successful PhD Students
Tim Dettmers
(ends 11:05 AM)
11:05 a.m.
Talk:
LMArena: An Open Platform for Crowdsourced AI benchmarks
Wei-Lin Chiang
(ends 11:25 AM)
11:25 a.m.
Talk:
Designing Models from the Hardware Up
Simran Arora
(ends 11:45 AM)
11:45 a.m.
Talk:
YPS - Talk by Beidi Chen
Beidi Chen
(ends 12:05 PM)
12:05 p.m.
Break:
Lunch Break
(ends 1:30 PM)
1:30 p.m.
Industry Lightning Talks [1:30-2:30]
(ends 2:30 PM)
2:30 p.m.
Panel Discussion:
Panel Discussion
(ends 3:30 PM)
4 p.m.
Poster Session and Reception - Young Professional Symposium [4:00-6:00]
(ends 6:00 PM)

TUE 13 MAY
8 a.m.
Registration and Check-in
(ends 4:00 PM)
8:30 a.m.
Opening Remarks:
Opening Remarks
(ends 8:45 AM)
8:45 a.m.
Session 1: LLM and Diffusion Model Serving [8:45-10:00]
FastTree: Optimizing Attention Kernel and Runtime for Tree-Structured LLM Inference
DiffServe: Efficiently Serving Text-to-Image Diffusion Models with Query-Aware Model Scaling
LeanAttention: Hardware-Aware Scalable Attention Mechanism for the Decode-Phase of Transformers
FlashInfer: Efficient and Customizable Attention Engine for LLM Inference Serving
Rethinking Key-Value Cache Compression Techniques for Large Language Model Serving
(ends 10:00 AM)
10 a.m.
Break:
Break
(ends 10:30 AM)
10:30 a.m.
Invited Talk:
An AI stack: from scaling AI workloads to evaluating LLMs
Ion Stoica
(ends 11:30 AM)
11:30 a.m.
Break:
Lunch Break
(ends 1:15 PM)
1:15 p.m.
Session 2: Parallel and Distributed Systems [1:15-2:40]
Context Parallelism for Scalable Million-Token Inference
GSplit: Scaling Graph Neural Network Training on Large Graphs via Split-Parallelism
Rubick: Exploiting Job Reconfigurability for Deep Learning Cluster Scheduling
PipeFill: Using GPUs During Bubbles in Pipeline-parallel LLM Training
AdaParse: An Adaptive Parallel PDF Parsing and Resource Scaling Engine
(ends 2:40 PM)
2:40 p.m.
Session 3: Quantization and Sparsity [2:40-4:00]
QServe:W4A8KV4 Quantization and System Co-design for Efficient LLM Serving
MiLo: Efficient Quantized MoE Inference with Mixture of Low-Rank Compensators
Enabling Unstructured Sparse Acceleration on Structured Sparse Accelerators
Radius: Range-based Gradient Sparsity for Large Foundation Model Pre-training
Self-Data Distillation for Recovering Quality in Pruned Large Language Models
(ends 4:00 PM)
4 p.m.
Break:
Break
(ends 4:45 PM)
4:45 p.m.
Session 4: Reliable and Scalable Systems [4:45-6:00]
Know Where You’re Uncertain When Planning with Multimodal Foundation Models: A Formal Framework
AIOpsLab: A Holistic Framework to Evaluate AI Agents for Enabling Autonomous Clouds
AI Metropolis: Scaling Large Language Model-based Multi-Agent Simulation with Out-of-order Execution
Interference-aware Edge Runtime Prediction with Conformal Matrix Completion
The Hidden Bloat in Machine Learning Systems
(ends 6:00 PM)
6 p.m.
Poster Session and Reception - Main Conference [6:00-8:00]
Posters -
QServe:W4A8KV4 Quantization and System Co-design for Efficient LLM Serving
SwiftVI: Time-Efficient Planning and Learning with MDPs
FastTree: Optimizing Attention Kernel and Runtime for Tree-Structured LLM Inference
ProtoRAIL: A Risk-cognizant Imitation Agent for Adaptive vCPU Oversubscription In the Cloud
Supply-Chain Attacks in Machine Learning Frameworks
VoLUT: Efficient Volumetric streaming enhanced by LUT-based super-resolution
Graph Learning at Scale: Characterizing and Optimizing Pre-Propagation GNNs
Know Where You’re Uncertain When Planning with Multimodal Foundation Models: A Formal Framework
Youmu: Efficient Columnar Data Pipeline for LLM Training
A Bring-Your-Own-Model Approach for ML-Driven Storage Placement in Warehouse-Scale Computers
LServe: Efficient Long-sequence LLM Serving with Unified Sparse Attention
DiffServe: Efficiently Serving Text-to-Image Diffusion Models with Query-Aware Model Scaling
LeanAttention: Hardware-Aware Scalable Attention Mechanism for the Decode-Phase of Transformers
Training Ultra Long Context Language Model with Fully Pipelined Distributed Transformer
Lightweight Software Kernels and Hardware Extensions for Efficient Sparse Deep Neural Networks on Microcontrollers
MiLo: Efficient Quantized MoE Inference with Mixture of Low-Rank Compensators
FedProphet: Memory-Efficient Federated Adversarial Training via Robust and Consistent Cascade Learning
Efficient On-Device Machine Learning with a Biologically-Plausible Forward-Only Algorithm
FLStore: Efficient Federated Learning Storage for non-training workloads
Enabling Unstructured Sparse Acceleration on Structured Sparse Accelerators
Optimizing LLM Queries in Relational Data Analytics Workloads
Marconi: Prefix Caching for the Era of Hybrid LLMs
FlexAttention: A Programming Model for Generating Fused Attention Variants.
FlashInfer: Efficient and Customizable Attention Engine for LLM Inference Serving
SampleAttention: Near-Lossless Acceleration of Long Context LLM Inference with Adaptive Structured Sparse Attention
Scaling Deep Learning Training with MPMD Pipeline Parallelism
Radius: Range-based Gradient Sparsity for Large Foundation Model Pre-training
Context Parallelism for Scalable Million-Token Inference
HyC-LoRA: Memory Efficient LoRA Fine-tuning with Hybrid Activation Compression
Seesaw: High-throughput LLM Inference via Model Re-sharding
ScaleFusion: Scalable Inference of Spatial-Temporal Diffusion Transformers for High-Resolution Long Video Generation
Efficient LLM Inference using Dynamic Input Pruning and Cache-Aware Masking
TurboAttention: Efficient attention approximation for high throughputs llm
AIOpsLab: A Holistic Framework to Evaluate AI Agents for Enabling Autonomous Clouds
GSplit: Scaling Graph Neural Network Training on Large Graphs via Split-Parallelism
TileLink: Generating Efficient Compute-Communication Overlapping Kernels using Tile-Centric Primitives
Self-Data Distillation for Recovering Quality in Pruned Large Language Models
COMET: Fine-grained Computation-communication Overlapping for Mixture-of-Experts
MAS-ATTENTION: MEMORY-AWARE STREAM PROCESSING FOR ATTENTION ACCELERATION ON RESOURCE-CONSTRAINED EDGE DEVICES
MEADOW: Memory-efficient Dataflow and Data Packing for Low Power Edge LLMs
AI Metropolis: Scaling Large Language Model-based Multi-Agent Simulation with Out-of-order Execution
Interference-aware Edge Runtime Prediction with Conformal Matrix Completion
APOLLO: SGD-like Memory, AdamW-level Performance
Lumos: Efficient Performance Modeling and Estimation for Large-scale LLM Training
ThunderServe: High-performance and Cost-efficient LLM Serving in Cloud Environments
Venn: Resource Management For Collaborative Learning Jobs
The Hidden Bloat in Machine Learning Systems
Balancing Pipeline Parallelism with Vocabulary Parallelism
Rethinking Key-Value Cache Compression Techniques for Large Language Model Serving
XGrammar: Flexible and Efficient Structured Generation Engine for Large Language Models
FlexInfer: Flexible LLM Inference with CPU Computations
On Distributed Larger-Than-Memory Subset Selection With Pairwise Submodular Functions
Rubick: Exploiting Job Reconfigurability for Deep Learning Cluster Scheduling
SOLA: Optimizing SLO Attainment for Large Language Model Serving with State-Aware Scheduling
NEO: Saving GPU Memory Crisis with CPU Offloading for Online LLM Inference
PipeFill: Using GPUs During Bubbles in Pipeline-parallel LLM Training
AdaParse: An Adaptive Parallel PDF Parsing and Resource Scaling Engine
ReaL: Efficient RLHF Training of Large Language Models with Parameter Reallocation
SparseTransX: Efficient Training of Translation-Based Knowledge Graph Embeddings Using Sparse Matrix Operations
LAVA: Lifetime-Aware VM Allocation with Learned Distributions and Adaptation to Mispredictions
Photon: Federated LLM Pre-Training
(ends 8:00 PM)

WED 14 MAY
8 a.m.
Registration and Check-in
(ends 4:00 PM)
8:30 a.m.
Session 5: LLM training and fine-tuning [8:30-10:00]
Youmu: Efficient Columnar Data Pipeline for LLM Training
Training Ultra Long Context Language Model with Fully Pipelined Distributed Transformer
HyC-LoRA: Memory Efficient LoRA Fine-tuning with Hybrid Activation Compression
APOLLO: SGD-like Memory, AdamW-level Performance
Lumos: Efficient Performance Modeling and Estimation for Large-scale LLM Training
ReaL: Efficient RLHF Training of Large Language Models with Parameter Reallocation
(ends 10:00 AM)
10 a.m.
Break:
Break
(ends 10:30 AM)
10:30 a.m.
Invited Talk:
Hardware-aware training and inference for large-scale AI
Animashree Anandkumar
(ends 11:30 AM)
11:30 a.m.
Break:
Lunch Break
(ends 1:15 PM)
1:15 p.m.
Session 6: Edge and Cloud Systems [1:15-2:40]
SwiftVI: Time-Efficient Planning and Learning with MDPs
ProtoRAIL: A Risk-cognizant Imitation Agent for Adaptive vCPU Oversubscription In the Cloud
A Bring-Your-Own-Model Approach for ML-Driven Storage Placement in Warehouse-Scale Computers
Efficient On-Device Machine Learning with a Biologically-Plausible Forward-Only Algorithm
Optimizing LLM Queries in Relational Data Analytics Workloads
(ends 2:40 PM)
2:40 p.m.
Session 7: Quantization and Sparsity [2:40-4:00]
LServe: Efficient Long-sequence LLM Serving with Unified Sparse Attention
Lightweight Software Kernels and Hardware Extensions for Efficient Sparse Deep Neural Networks on Microcontrollers
SampleAttention: Near-Lossless Acceleration of Long Context LLM Inference with Adaptive Structured Sparse Attention
Efficient LLM Inference using Dynamic Input Pruning and Cache-Aware Masking
SparseTransX: Efficient Training of Translation-Based Knowledge Graph Embeddings Using Sparse Matrix Operations
(ends 4:00 PM)
4 p.m.
Break:
Break
(ends 4:45 PM)
4:45 p.m.
Session 8: LLM and Diffusion Model Serving [4:45-6:00]
Seesaw: High-throughput LLM Inference via Model Re-sharding
ScaleFusion: Scalable Inference of Spatial-Temporal Diffusion Transformers for High-Resolution Long Video Generation
TurboAttention: Efficient attention approximation for high throughputs llm
FlexInfer: Flexible LLM Inference with CPU Computations
SOLA: Optimizing SLO Attainment for Large Language Model Serving with State-Aware Scheduling
(ends 6:00 PM)
6 p.m.
Poster Session - Optional [6:00-8:00]
(ends 8:00 PM)

THU 15 MAY
8 a.m.
Registration and Check-in
(ends 3:00 PM)
8:30 a.m.
Session 9: Parallel and Distributed Systems [8:30-10:00]
Scaling Deep Learning Training with MPMD Pipeline Parallelism
TileLink: Generating Efficient Compute-Communication Overlapping Kernels using Tile-Centric Primitives
COMET: Fine-grained Computation-communication Overlapping for Mixture-of-Experts
Balancing Pipeline Parallelism with Vocabulary Parallelism
On Distributed Larger-Than-Memory Subset Selection With Pairwise Submodular Functions
(ends 10:00 AM)
10 a.m.
Break:
Break
(ends 10:30 AM)
10:30 a.m.
Invited Talk:
Responsible Finetuning of Large Language Models
Ling Liu
(ends 11:30 AM)
11:30 a.m.
Break:
Lunch Break
(ends 1:15 PM)
1:15 p.m.
Session 10: LLM and Diffusion Model Serving [1:15-2:40]
Marconi: Prefix Caching for the Era of Hybrid LLMs
FlexAttention: A Programming Model for Generating Fused Attention Variants.
ThunderServe: High-performance and Cost-efficient LLM Serving in Cloud Environments
XGrammar: Flexible and Efficient Structured Generation Engine for Large Language Models
NEO: Saving GPU Memory Crisis with CPU Offloading for Online LLM Inference
(ends 2:40 PM)
2:40 p.m.
Session 11: Federated Learning [2:40-4:00]
FedProphet: Memory-Efficient Federated Adversarial Training via Robust and Consistent Cascade Learning
FLStore: Efficient Federated Learning Storage for non-training workloads
MAS-ATTENTION: MEMORY-AWARE STREAM PROCESSING FOR ATTENTION ACCELERATION ON RESOURCE-CONSTRAINED EDGE DEVICES
Venn: Resource Management For Collaborative Learning Jobs
Photon: Federated LLM Pre-Training
(ends 4:00 PM)
4 p.m.
Break:
Break
(ends 4:45 PM)
4:45 p.m.
Session 12: Edge and Cloud Systems [4:45-6:00]
Supply-Chain Attacks in Machine Learning Frameworks
VoLUT: Efficient Volumetric streaming enhanced by LUT-based super-resolution
Graph Learning at Scale: Characterizing and Optimizing Pre-Propagation GNNs
MEADOW: Memory-efficient Dataflow and Data Packing for Low Power Edge LLMs
LAVA: Lifetime-Aware VM Allocation with Learned Distributions and Adaptation to Mispredictions
(ends 6:00 PM)
6 p.m.
Closing Remarks:
Closing Remarks
(ends 6:05 PM)