Pin-Yu Chen

Zeroth-order optimization offers memory efficiency for large language models in NLP tasks.

A fresh method to improve text-to-image models with efficiency and quality.

A new benchmark to improve ASR accuracy using language models.

A smart system to recognize multiple languages without prior training.

New methods enhance watermarking in large language models while preserving text quality.

New methods enhance accuracy in noisy speech recognition using large language models.

Learn how model reprogramming enhances machine learning without heavy adjustments.

Larimar enhances large language models with real-time memory updates.

A new method integrates acoustic information into language models for better speech recognition.

Quantum computing boosts machine learning efficiency and capabilities through innovative collaboration.

Discover a new method for efficient neural architecture search through graph-based evaluations.

This article examines the risks of fine-tuning language models for safety.

Analyzing the factors affecting alignment in large language models.

A new system aims to enhance sepsis prediction despite challenges in missing data.

A fresh method reveals how vision models organize and understand images.

Addressing security risks in generative AI through red and blue teaming.

Examining the safety risks of using language models for drone operations.

Researchers develop MergeAlign to make AI safer without losing expertise.

Exploring the blend of quantum computing and machine learning for future innovations.

Discover how quantum computing and machine learning intersect to transform various fields.

ICER framework tests safety measures in text-to-image models effectively.

A new dataset improves how models perceive color and context.