edge devices
Computational hardware at the periphery of a network that performs data processing locally rather than relying on a centralized cloud, enabling faster response times and reduced latency for AI applications, particularly in IoT and mobile environments.
- CoreGuard: Safeguarding Foundational Capabilities of LLMs Against Model Stealing in Edge Deployment
- DynaNav: Dynamic Feature and Layer Selection for Efficient Visual Navigation
- LoRO: Real-Time on-Device Secure Inference for LLMs via TEE-Based Low Rank Obfuscation
- Mellow: a small audio language model for reasoning
- PMQ-VE: Progressive Multi-Frame Quantization for Video Enhancement
- PUO-Bench: A Panel Understanding and Operation Benchmark with A Privacy-Preserving Framework
- Resource-Constrained Federated Continual Learning: What Does Matter?
- Towards Predicting Any Human Trajectory In Context
- Understanding outer learning rates in Local SGD