相关论文: Power efficient carry propagate adder
Experiments have shown that the light-emission efficiency of indium gallium nitride (InGaN) light-emitting diodes improves with increasing indium concentration. It is widely thought that compositional fluctuations due to indium…
Existing modeling approaches for long-duration energy storage (LDES) are often based either on an oversimplified representation of power system operations or limited representation of storage technologies, e.g., evaluation of only a single…
In this paper, we propose a 'full-stack' solution to designing high capacity and low latency on-chip cache hierarchies by starting at the circuit level of the hardware design stack. First, we propose a novel Gain Cell (GC) design using…
Discrete diffusion models have achieved strong empirical performance in text and other symbolic domains, but, especially for uniform-rate models, they often require many steps to generate a single sample. Existing acceleration methods…
This paper proposes a co-optimization generation and distribution planning model in microgrids in which simultaneous investment in generation, i.e., distributed generation (DG) and distributed energy storage (DES), and distribution, i.e.,…
Caching is crucial for enabling high-throughput networks for data intensive applications. Traditional caching technology relies on DRAM, as it can transfer data at a high rate. However, DRAM capacity is subject to contention by most system…
It is shown that semiconductor nanoscale resonators with extreme dielectric confinement accelerate the diffusion of electron-hole pairs excited by nonlinear absorption. These novel cavity designs may lead to optical switches with superior…
The study proposes an ultra-thin back side illuminated (BSI) and top-illuminated, Ge on Si photodetector (PD), for 1 to 1.4 microns wavelength range. The Ge thickness of 350 nm allows us to achieve high-speed performance at >60 GHz, while…
In modern computing units, division operations are generally slower than other arithmetic operations and require more resources, such as area and power, than multiplication. To reduce the delay, fast division algorithms use an initial…
Advanced electronic device technologies require a faster operation and smaller average power consumption, which are the most important parameters in very large scale integrated circuit design. The conventional Complementary Metal-Oxide…
Portable devices like smartphones, tablets, wearable electronic devices, medical implants, wireless sensor nodes, and Internet-of-Things (IoT) devices have tremendous constraints on their energy consumption. Adding more functionalities onto…
This paper presents an energy-efficient stable diffusion processor for text-to-image generation. While stable diffusion attained attention for high-quality image synthesis results, its inherent characteristics hinder its deployment on…
Recently, 3D Gaussian Splatting (3DGS) has emerged as a mainstream rendering technique due to its photorealistic quality and low latency. However, processing massive numbers of non-contributing Gaussian points introduces significant…
Diffusion models represent a powerful family of generative models widely used for image and video generation. However, the time-consuming deployment, long inference time, and requirements on large memory hinder their applications on…
While classifier-free guidance (CFG) is essential for conditional diffusion models, it doubles the number of neural function evaluations (NFEs) per inference step. To mitigate this inefficiency, we introduce adapter guidance distillation…
A novel design of a grating-based optical pulse compressor is proposed. The proposed compressor provides a large group delay dispersion while keeping the compressor linear size small. The design of the proposed compressor is based on a…
Massive MIMO systems are typically designed assuming linear power amplifiers (PAs). However, PAs are most energy efficient close to saturation, where non-linear distortion arises. For conventional precoders, this distortion can coherently…
Diffusion model deployment has been suffering from high energy consumption and inference latency despite its superior performance in visual generation tasks. Dynamic voltage and frequency scaling (DVFS) offers a promising solution to…
Distributed massive MIMO (D-mMIMO) has been considered for future networks as it holds the potential to offer superior capacity while enabling energy savings in the network. A D-mMIMO system has multiple arrays. Optimizing the locations of…
Transforming large pre-trained low-resolution diffusion models to cater to higher-resolution demands, i.e., diffusion extrapolation, significantly improves diffusion adaptability. We propose tuning-free CutDiffusion, aimed at simplifying…