Ya Li

Domain Generalization via Conditional Invariant Representation

2018 · arXiv (Cornell University)

Domain generalization aims to apply knowledge gained from multiple labeled source domains to unseen target domains. The main difficulty comes from the dataset bias: training data and test data have different distributions, and the training …

Detection of cigarette smoke using a fiber membrane filmed with carbon nanoparticles and a fractal current law

2020 · Thermal Science

Cigarette smoke will cause irreversible damage to human health and living environment. It is especially important to accurately detect and effectively prevent cigarette smoke. This paper fabricates a fiber membrane filmed with carbon nanopowders as …

A Fast Granular-Ball-Based Density Peaks Clustering Algorithm for Large-Scale Data

2023 · IEEE Transactions on Neural Networks and Learning Systems

Density peaks clustering algorithm (DP) has difficulty in clustering large-scale data, because it requires the distance matrix to compute the density and -distance for each object, which has time complexity. Granular ball (GB) is a …

CJT: Boosting the Transferability of Adversarial Attacks with Color Jitter Transformation

2023

The transferability of adversarial examples in black-box attacks is crucial for the performance of the attack. However, the gap between models weakens the attack performance when it is transferred from the surrogate model to the …

Retrieval Augmented Generation in Prompt-based Text-to-Speech Synthesis with Context-Aware Contrastive Language-Audio Pretraining

2024 · arXiv (Cornell University)

Recent prompt-based text-to-speech (TTS) models can clone an unseen speaker using only a short speech prompt. They leverage a strong in-context ability to mimic the speech prompts, including speaker style, prosody, and emotion. Therefore, the …