Wei Zou

Asymmetric quantization in hardware accelerator

2023

This paper presents an efficient implementation of asymmetric quantization in hardware accelerator for deep learning applications. In this work, we show that asymmetric quantization provides better accuracy performance in AI inferencing with the same amount …

Extend Adversarial Policy Against Neural Machine Translation via Unknown Token

2025 · arXiv (Cornell University)

Generating adversarial examples contributes to mainstream neural machine translation~(NMT) robustness. However, popular adversarial policies are apt for fixed tokenization, hindering its efficacy for common character perturbations involving versatile tokenization. Based on existing adversarial generation via …

Understanding the Modality Gap: An Empirical Study on the Speech-Text Alignment Mechanism of Large Speech Language Models

2025 · arXiv (Cornell University)

End-to-end Large Speech Language Models (LSLMs) have demonstrated impressive conversational generation abilities, yet consistently fall short of traditional pipeline systems on semantic understanding benchmarks. In this work, we reveal through systematic experimentation that although LSLMs …

Improving Transformer-based Speech Recognition Using Unsupervised Pre-training

2019 · arXiv (Cornell University)

Speech recognition technologies are gaining enormous popularity in various industrial applications. However, building a good speech recognition system usually requires large amounts of transcribed data, which is expensive to collect. To tackle this problem, an …