2026.03
CRANE was open-sourced for direct Apple Neural Engine inference
CRANE is now open-sourced for direct Apple Neural Engine inference without Core ML.
Ph.D. in Computer Sciences
Yilong Li
University of Wisconsin-Madison, advised by Prof. Suman Banerjee
I am a systems researcher who builds both hardware and software for intelligent edge systems. My work focuses on on-device AI, reinforcement-learning post-training for efficient edge intelligence, and wireless sensing systems.
I build the full stack: custom wearable and embedded hardware, accelerator-aware runtimes, sensing platforms, model adaptation pipelines, and post-training methods that make AI run under real-world compute, memory, energy, and privacy constraints. My work has appeared in MobiCom, ICLR, NSDI, and SenSys.
Recent Momentum
News / Updates
2026.01
NanoMind was accepted to ICLR 2026
Our work on efficient multimodal inference for battery-powered small devices was accepted to ICLR 2026.
2025.06
Distributed MIMO Radars was accepted to MobiCom 2025
Our work on scalable biometric sensing in the wild through distributed MIMO radars was accepted to MobiCom 2025.
2025.01
PalmBench was accepted to ICLR 2025
PalmBench, our benchmark of compressed large language models on mobile platforms, was accepted to ICLR 2025.
Current Directions
Ongoing Research
Selected projects below show the full stack: custom hardware, accelerator-aware software, post-training and model adaptation, and wireless sensing.
Theme 01
On-Device AI Systems
Building battery-powered multimodal devices and local inference systems that combine sensors, embedded hardware, and accelerator-aware software.
Theme 02
Model Efficiency and RL Post-Training
Using quantization, cross-accelerator execution, and reinforcement learning to adapt models for edge devices with tight compute and memory budgets.
Theme 03
Wireless Sensing Systems
Designing UWB, mmWave, and distributed MIMO sensing systems that combine RF hardware, embedded software, and learning algorithms.
Selected Project
EMBER / StoreAgent: RL-Trained Memory Policies for Efficient Edge AI
Model-level work on reinforcement-learning post-training for long-horizon agents. EMBER and StoreAgent train memory policies that decide what evidence to retain, how to index it, and how to recall it under strict context and memory budgets.
Explore project
Selected Project
MEDUSA / Gemini: Wireless Sensing from RF Hardware to Learning Systems
Wireless sensing systems that span UWB, mmWave, and distributed MIMO radar. These projects combine RF hardware, synchronization, embedded software, and learning algorithms for robust vital-sign, motion, and integrated sensing-communication applications.
Explore project
Selected Project
Split to Fit / CRANE: Cross-Accelerator Software for Efficient Edge AI
Systems software for running multimodal models on heterogeneous edge accelerators. Split to Fit maps vision and language components across NPUs and GPUs with hybrid quantization, while CRANE exposes direct Apple Neural Engine execution without Core ML.
Explore project
Selected Project
Virgile: A Multimodal Visual Memory Assistant with Persistent Object and Face Recognition on Edge Devices
A wearable multimodal assistant built on the NanoMind hardware platform. The project combines custom sensors, embedded hardware, accelerator-aware software, and local vision-language inference to support persistent object and face recognition on edge devices.
Explore projectSelected Output
Selected Publications
Recent work on efficient multimodal inference, mobile AI benchmarking, and wireless sensing systems.
Tiny but Mighty: A Software-Hardware Co-Design Approach for Efficient Multimodal Inference on Battery-Powered Small Devices
Large Multimodal Models (LMMs) are inherently modular, consisting of vision and audio encoders, projectors, and large language models. Yet, they are almost always executed monolithically, which underutilizes the heter...
Scalable Biometric Sensing in the Wild through Distributed MIMO Radars
Radar-based techniques for detecting vital signs have shown promise for continuous contactless vital sign sensing and healthcare applications. However, real-world indoor environments face significant challenges for ex...
PalmBench: A Comprehensive Benchmark of Compressed Large Language Models on Mobile Platforms
Deploying large language models (LLMs) locally on mobile devices is advantageous in scenarios where transmitting data to remote cloud servers is either undesirable due to privacy concerns or impractical due to network...
Gemini: Integrating Full-fledged Sensing upon Millimeter Wave Communications
Integrating millimeter wave (mmWave)technology in both communication and sensing is promising as it enables the reuse of existing spectrum and infrastructure without draining resources. Most existing systems piggyback...
LiveTag: Sensing Human-Object Interaction through Passive Chipless WiFi Tags
Many types of human activities involve interaction with passive objects. Thus, by wirelessly sensing human interaction with them, one can infer activities at a fine resolution, enabling a new wave of ubiquitous comp...