Yong Li
where else to find them
no profile links on file
what they talk about
all 13 in the filters
where they keep going back
no series more than once — all 11 in the filters
on a stage with
all 114 in the filters
who talks about the same things
nobody close enough yet
| title | conference | date |
|---|---|---|
| Tessera: A Holistic Pipeline Parallelism Framework for Trillion-Parameter Heterogeneous | 20th USENIX Symposium on Operating Systems Design and Implementation (OSDI '26) | 2026-07-13 |
Tessera is a pipeline-parallel framework that profiles fine-grained, per-pair overlap schedules and uses those measured post-overlap costs to cut stages, then… | ||
| TrainMover: An Interruption-Resilient Runtime for ML Training | 20th USENIX Symposium on Operating Systems Design and Implementation (OSDI '26) | 2026-07-13 |
TrainMover is an ML training runtime that prepares standby GPUs ahead of interruptions by pre-recording output tensors and running communication-free sandboxed… | ||
| EROICA: Online Performance Troubleshooting for Large-scale Model Training | NSDI '26 | 2026-05-04 |
EROICA is an online profiler that triggers only on iteration slowdowns, profiles all workers simultaneously, extracts beta, mu, sigma fingerprints for… | ||
| Evolution of Aegis: Fault Diagnosis for AI Model Training Service in Production | NSDI '25 | 2025-04-28 |
Aegis is a production fault diagnosis system for AI model training that evolves from enhanced legacy tools to a customized collective communication library… | ||
| Llumnix: Dynamic Scheduling for Large Language Model Serving | 18th USENIX Symposium on Operating Systems Design and Implementation (OSDI '24) | 2024-07-10 |
Llumnix implements live migration of per-request KV caches to reschedule running LLM inferences across instances with near-zero downtime, using a distributed… | ||
| Legion: Automatically Pushing the Envelope of Multi-GPU System for Billion-Scale... | USENIX ATC '23 | 2023-07-10 |
Claims an end-to-end multi-GPU system that trains billion‑node GNNs on a single machine by combining hierarchical graph partitioning to minimize cache… | ||
| Whale: Efficient Giant Model Training over Heterogeneous GPUs | USENIX ATC '22 | 2022-07-11 |
Whale is a distributed training framework that adds two model-annotation primitives, replicate and split, to transform a local DAG into a distributed plan and… | ||
| MLaaS in the Wild: Workload Analysis and Scheduling in Large-Scale Heterogeneous GPU... | 19th USENIX Symposium on Networked Systems Design and Implementation (NSDI '22) | 2022-04-04 |
A two-month Alibaba MLaaS trace covers over 6,700 GPUs across ~1,800 nodes with 1.2M tasks, 7.5M instances and shows concrete pain points: 85% gang-scheduled… | ||
| EPMC Gets Knowledge in Multi-Agent Systems | VMCAI 2022 | 2022-01-16 |
EPMC is a small Java core plus plugins probabilistic model checker that the authors extend to EPMC-petl to handle probabilistic epistemic temporal logic on… | ||
| When Sharing Economy Meets IoT: Towards Fine-grained Urban Air quality Monitoring through Mobile ... | UbiComp 2021 | 2021-09-21 |
UbiAir mounts an IoT sensor box on shared bikes, collects >100k PM2.5 readings from 36 students over 18 days, and splits the area into fine grids. It applies… | ||
| Who Will Survive and Revive Undergoing the Epidemic: Analyses about POI Visit Behavior inWuhan ... | UbiComp 2021 | 2021-09-21 |
Uses a >140,000-POI check-in dataset with two parallel timelines to compare 2019 and 2020 and extracts trends by phase. Visits fall more than 60% on average… | ||
| Passive Health Monitoring Using Large Scale Mobility Data | UbiComp 2021 | 2021-09-21 |
Uses mobility traces from 2112 users collected for about two months, extracts 16 features (radius of gyration, displacement std, visit entropy, POI counts)… | ||
| Prescriptively Benchmarking Kubernetes System and Application Using K-Bench | KubeCon + CloudNativeCon North America 2020 Virtual | 2020-11-17 |
K-Bench is a configurable framework that provisions Kubernetes resources from YAML, runs 1000-pod startup test style workloads and reports… | ||
| Learning from Hometown and Current City: Cross-city POI Recommendation via Interest Drift and ... | UbiComp / ISWC 2020 | 2020-09-12 |
Proposes PR-UIDT, a matrix-factorization framework that splits each user vector into a city-independent part and a city-dependent part and represents each POI… | ||
| Personalized Context-aware Collaborative Online Activity Prediction | UbiComp / ISWC 2020 | 2020-09-12 |
Builds a user-location-time-activity 4D-tensor and a location-time-POI 3D-tensor, uses Telecom app logs (7 days, ~2000 apps) and Check-in data (3 months, 16… | ||
| PMF: A Privacy-preserving Human Mobility Prediction Framework via Federated Learning | UbiComp / ISWC 2020 | 2020-09-12 |
Proposes a mobility predictor that keeps raw trajectories on-device and only exchanges model weights, combining federated learning with differential privacy… | ||
| nothing matches | ||