1

A^2: Characterizing and Mitigating Imbalanced Reclaim in MGLRU via Active Aging at Production Scale
A^2: Characterizing and Mitigating Imbalanced Reclaim in MGLRU via Active Aging at Production Scale. Accepted to the ACM SIGOPS Annual Technical Conference (ATC 2026).
TensorGPT: Towards Generative Tensor Program Optimization for Heterogeneous Edge Devices
TensorGPT: Towards Generative Tensor Program Optimization for Heterogeneous Edge Devices. Conditionally accepted to ACM SenSys 2027.
Surviving the Impossible Trinity: Revisiting CPU Scheduling Problem on Modern COTS Mobile Devices
Surviving the Impossible Trinity: Revisiting CPU Scheduling Problem on Modern COTS Mobile Devices. Published in USENIX OSDI 2026. July 13–15, 2026, Seattle, WA, USA.
LLM-Flex: Adaptive Heterogeneous Federated Learning Based on LLM in Uncertain Scenarios
LLM-Flex: Adaptive Heterogeneous Federated Learning Based on LLM in Uncertain Scenarios. Published in CSCWD 2026.
PDF
DigimonGPT: An Evolvable Agent with Hierarchical Human-like Memory for Video Question Answering
DigimonGPT: An Evolvable Agent with Hierarchical Human-like Memory for Video Question Answering. Published in AAAI'26. Jan. 20 – 27, 2026, Singapore.
TensorShield: Safeguarding On-Device Inference by Shielding Critical DNN Tensors with TEE
TensorShield: Safeguarding On-Device Inference by Shielding Critical DNN Tensors with TEE. Published in ACM CCS 2025. October 13-17, 2025.
MobiLoRA: Accelerating LoRA-based LLM Inference on Mobile Devices via Context-aware KV Cache Optimization
MobiLoRA: Accelerating LoRA-based LLM Inference on Mobile Devices via Context-aware KV Cache Optimization. Published in ACL 2025. Vienna, Austria. July 27–August 1st, 2025.
InfiniCL: Elastic Continual Learning for Resource-Constrained Edge Devices
InfiniCL: Elastic Continual Learning for Resource-Constrained Edge Devices. Published in ACM/IEEE IWQoS 2025. Gold Coast, Australia, July 2-4, 2025.
InfScaler: Enabling Efficient ML Inference Serving on Multi-Accelerator Edge Devices via Asymmetric Auto-Scaling
InfScaler: Enabling Efficient ML Inference Serving on Multi-Accelerator Edge Devices via Asymmetric Auto-Scaling. Published in IEEE/ACM DAC 2025. San Francisco, CA, USA. June 22-25, 2025.
PDF
Jolteon: Unleashing the Promise of Serverless for Serverless Workflows
Jun 6, 2024
PDF Slides