Large Language Models (LLMs) have transformed natural language processing, but face significant challenges in widespread deployment due to their high runtime cost. In this paper, we introduce SeedLM, a novel post-training compression method that uses seeds of a pseudo-random generator to encode and compress model weights. Specifically, for each block of weights, we find a seed that is fed into a Linear Feedback Shift Register (LFSR) during inference to efficiently generate a random matrix. This matrix is then linearly combined with compressed coefficients to reconstruct the weight block. SeedLM reduces memory access and leverages idle compute cycles during inference, effectively speeding up memory-bound tasks by trading compute for fewer memory accesses. Unlike state-of-the-art methods that rely on calibration data, our approach is data-free and generalizes well across diverse tasks. Our experiments with Llama3 70B, which is particularly challenging, show zero-shot accuracy retention at 4- and 3-bit compression to be on par with or better than state-of-the-art methods, while maintaining performance comparable to FP16 baselines. Additionally, FPGA-based tests demonstrate that 4-bit SeedLM, as model size increases, approaches a 4x speed-up over an FP16 Llama 2/3 baseline.

Wireless video streaming has traditionally been considered an extremely power-hungry operation. Existing approaches optimize the camera and communication modules individually to minimize their power consumption. However, designing a video streaming device requires power-consuming hardware components and video CODEC algorithms which makes battery-free video streaming currently infeasible. Existing RF-powered wireless camera prototypes require extensive duty-cycling on the order of tens of minutes, to capture, process and communicate a single frame. Self-powered cameras can capture an image once every few seconds, but do not have the capability to stream video wirelessly.

We consider the problem of continuous computer-vision based analysis of video streams from mobile cameras over extended periods. Given high computational demands, general visual processing must currently be offloaded to the cloud. To reduce mobile battery and bandwidth consumption, recent proposals offload only "interesting" video frames, discarding the rest. However, determining what to discard is itself typically a power-hungry computer vision calculation, very often well beyond what most mobile devices can afford on a continuous basis. We present the Glimpse system, a re-design of the conventional mobile video processing pipeline to support such "early discard" flexibly, efficiently and accurately. Glimpse is a novel architecture that gates wearable vision using low-power vision modalities. Our proposed architecture adds novel sensing, processing, algorithmic and programming-system components to the camera pipeline to this end. We present a complete implementation and evaluation of our design. In common settings, Glimpse reduces mobile power and data usage by more than one order of magnitude relative to earlier designs, and moves continuous vision on lightweight wearables to the realm of the practical.

We present the first power over Wi-Fi system that delivers power to low-power sensors and devices and works with existing Wi-Fi chipsets. Specifically, we show that a ubiquitous part of wireless communication infrastructure, the Wi-Fi router, can provide far field wireless power without significantly compromising the network's communication performance. Building on our design, we prototype battery-free temperature and camera sensors that we power with Wi-Fi at ranges of 20 and 17 feet respectively. We also demonstrate the ability to wirelessly trickle-charge nickel---metal hydride and lithium-ion coin-cell batteries at distances of up to 28 feet. We deploy our system in six homes in a metropolitan area and show that it can successfully deliver power via Wi-Fi under real-world network conditions without significantly degrading network performance.

RFID sensor networks perpetually stream sensor data without batteries. Cameras are power hungry but provide richer information than conventional sensor network nodes. Battery-free, RF-powered camera sensor nodes combine many of the attractive features of RFID sensor networks with those of cameras. However, prior battery-free cameras have no notion of 3D location, which is desirable for creating large scale networks of battery free cameras.