GitHub

Revolutionizing Reinforcement Learning Framework for Diffusion Large Language Models

We propose TraceRL, a trajectory-aware reinforcement learning method for diffusion language models, which demonstrates the best performance among RL approaches for DLMs. We also introduce a ...
Semiconductor Engineering

Addressing Critical Tradeoffs In NPU Design

AI/ML are driving a steep ramp in neural processing unit (NPU) design activity for everything from data centers to edge ...
IEEE

A novel technique for Software Fault Injection Testing through FPGA code instrumentation ...

Abstract: Software Fault Injection Testing (SFIT) is a technique used in verification & validation (V&V) in order to test the error handling logics in the software on ...
IEEE

FPGA implementation of a Single Pass Connected Components Algorithm

Abstract: The classic connected components labelling algorithm requires two passes through an image. This paper presents an algorithm that allows the connected components to be analysed in a single ...