55 citations · 66 across the 10 of their papers we have counts for
10 papers
From Architecture to Output: Structural Origins of Hallucination in Large Language Models and the Amplifying Role of Data
Md. Rejaul Korim Sadi, Toufiqur Rahman Tasin, Golam Mostofa Naeem
Large language models produce fluent, confident, factually wrong output. Existing taxonomies classify these failures by output type -- intrinsic versus extrinsic, faithfulness vers…
Chiplet-Gym: Optimizing Chiplet-based AI Accelerator Design with Reinforcement Learning
Kaniz Mishty, Mehdi Sadi
Modern Artificial Intelligence (AI) workloads demand computing systems with large silicon area to sustain throughput and competitive performance. However, prohibitive manufacturing…
SNNLP: Energy-Efficient Natural Language Processing Using Spiking Neural Networks
R. Alexander Knipper, Kaniz Mishty, Mehdi Sadi +1
As spiking neural networks receive more attention, we look toward applications of this computing paradigm in fields other than computer vision and signal processing. One major fiel…
System and Design Technology Co-optimization of SOT-MRAM for High-Performance AI Accelerator Memory System
Kaniz Mishty, Mehdi Sadi
SoCs are now designed with their own AI accelerator segment to accommodate the ever-increasing demand of Deep Learning (DL) applications. With powerful MAC engines for matrix multi…
Attacking Deep Learning AI Hardware with Universal Adversarial Perturbation
Mehdi Sadi, B. M. S. Bahar Talukder, Kaniz Mishty +1
Universal Adversarial Perturbations are image-agnostic and model-independent noise that when added with any image can mislead the trained Deep Convolutional Neural Networks into th…
Designing Efficient and High-performance AI Accelerators with Customized STT-MRAM
Kaniz Mishty, Mehdi Sadi
In this paper, we demonstrate the design of efficient and high-performance AI/Deep Learning accelerators with customized STT-MRAM and a reconfigurable core. Based on model-driven d…
True Random Number Generation using Latency Variations of Commercial MRAM Chips
Farah Ferdaus, B. M. S. Bahar Talukder, Mehdi Sadi +1
The emerging magneto-resistive RAM (MRAM) has considerable potential to become a universal memory technology because of its several advantages: unlimited endurance, lower read/writ…
Special Session: Reliability Analysis for ML/AI Hardware
Shamik Kundu, Kanad Basu, Mehdi Sadi +4
Artificial intelligence (AI) and Machine Learning (ML) are becoming pervasive in today's applications, such as autonomous vehicles, healthcare, aerospace, cybersecurity, and many c…
Yield Loss Reduction and Test of AI and Deep Learning Accelerators
Mehdi Sadi, Ujjwal Guin
With data-driven analytics becoming mainstream, the global demand for dedicated AI and Deep Learning accelerator chips is soaring. These accelerators, designed with densely packed…
Hardware Trojan Detection through Information Flow Security Verification
Adib Nahiyan, Mehdi Sadi, Rahul Vittal +3
Semiconductor design houses are increasingly becoming dependent on third party vendors to procure intellectual property (IP) and meet time-to-market constraints. However, these thi…