Related papers: Optimized Vectorization Implementation of CRYSTALS…
Digital Computing-in-Memory (DCIM) is an innovative technology that integrates multiply-accumulation (MAC) logic directly into memory arrays to enhance the performance of modern AI computing. However, the need for customized memory cells…
We describe a set of lower-level abstractions to improve performance on modern large scale heterogeneous systems. These provide portable access to system- and hardware-dependent features, automatically apply dynamic optimizations at run…
Various gradient compression schemes have been proposed to mitigate the communication cost in distributed training of large scale machine learning models. Sign-based methods, such as signSGD, have recently been gaining popularity because of…
It is notably challenging to design an efficient and secure signature scheme based on error-correcting codes. An approach to build such signature schemes is to derive it from an identification protocol through the Fiat-Shamir transform. All…
Digital Decimation process plays an important task in a communication system. It mostly is applied in transceiver when the frequency reduction is required. However, the decimation process for sigma-delta modulator is considered in this…
The emerging mobile devices in this era of internet-of-things (IoT) require a dedicated processor to enable computationally intensive applications such as neuromorphic computing and signal processing. Vector-by-matrix multiplication (VMM)…
Vector search systems are indispensable in large language model (LLM) serving, search engines, and recommender systems, where minimizing online search latency is essential. Among various algorithms, graph-based vector search (GVS) is…
Diffusion models produce high-quality synthetic data but suffer from slow inference. We propose 3D Variable-Step Denoising Diffusion Probabilistic Model (VS-DDPM) a framework engineered to maintain generative quality while accelerating…
Inference-time scaling has emerged as an effective way to improve generative models at test time by using a verifier to score and select candidate outputs. A common choice is to employ Multimodal Large Language Models (MLLMs) as verifiers,…
A key factor in effective Self-Supervised learning (SSL) is preventing dimensional collapse, where higher-dimensional representation spaces ($R$) span a lower-dimensional subspace. Therefore, SSL optimization strategies involve guiding a…
This study evaluates numerical discretization methods for the Single Particle Model (SPM) used in electrochemical modeling. The methods include the Finite Difference Method (FDM), spectral methods, Pad\'e approximation, and parabolic…
We introduce the new AXU hash function decBRWHash, which is parameterised by the positive integer $c$ and is based on Bernstein-Rabin-Winograd (BRW) polynomials. Choosing $c>1$ gives a hash function which can be implemented using $c$-way…
Methods: Nine SSL methods spanning four pretext-task families were pretrained from scratch using the same 10{,}412 3D CT scans (1.89~M 2D axial slices) covering varied disease sites. The pretrained Swin Transformer encoder from each method…
Supervised learning with deep convolutional neural networks (DCNNs) has seen huge adoption in stereo matching. However, the acquisition of large-scale datasets with well-labeled ground truth is cumbersome and labor-intensive, making…
Privacy preservation is addressed for decentralized optimization, where $N$ agents cooperatively minimize the sum of $N$ convex functions private to these individual agents. In most existing decentralized optimization approaches,…
This paper introduces a new stochastic optimization method based on the regularized Fisher information matrix (FIM), named SOFIM, which can efficiently utilize the FIM to approximate the Hessian matrix for finding Newton's gradient update…
The augmented usage of deep learning-based models for various AI related problems are as a result of modern architectures of deeper length and the availability of voluminous interpreted datasets. The models based on these architectures…
An adaptive inference method for crossbar (AIDX) is presented based on an optimization scheme for adjusting the duration and amplitude of input voltage pulses. AIDX minimizes the long-term effects of memristance drift on artificial neural…
The computational complexity and efficiency of the approximate mode component synthesis (ACMS) method is investigated for the two-dimensional heterogeneous Helmholtz equations, aiming at the simulation of large but finite-size photonic…
In current computer architectures, data movement (from die to network) is by far the most energy consuming part of an algorithm (10pJ/word on-die to 10,000pJ/word on the network). To increase memory locality at the hardware level and reduce…