Nvidia shrinks LLM memory 20x without changing model weights

Nvidia's KV Cache Transform Coding (KVTC) compresses LLM key-value cache by 20x without model changes, cutting GPU memory costs and time-to-first-token by up to 8x for multi-turn AI applications.
Total Retail

How I Combined GPT With Classical Optimization to Make Retail AI 85% Faster

In large retail operations, category management teams spend significant time deciding which product goes onto which shelf and ...
IEEE

An Efficient Batch Bayesian Optimization Algorithm via Expected Coordinate Improvement

Abstract: Parallel Bayesian optimization is crucial for solving expensive black-box problems, yet batch acquisition strategies remain a challenge. To address this, we propose a novel parallel ...
IEEE

A Study on Multi-Objective Design Optimization Models and Algorithms for Reliability-Aware PCB Design

Abstract: Modern electronic devices demand ever-smaller, higher-performance printed circuit boards (PCBs), yet miniaturization and complex service environments exacerbate failure risks. We first ...