Rui Zhao

An Effective Crop-Paste Pipeline for Few-shot Object Detection

2023 · arXiv (Cornell University)

Few-shot object detection (FSOD) aims to expand an object detector for novel categories given only a few instances for training. However, detecting novel categories with only a few samples usually leads to the problem of …

Scalable and Privacy-Preserving Synthetic Data Generation on Decentralised Web

2023 · arXiv (Cornell University)

Data on the Web has fueled much of the recent progress in AI. As more high-quality data becomes difficult to access, synthetic data is emerging as a promising solution for privacy-friendly data release and complementing …

Data Pruning via Separability, Integrity, and Model Uncertainty-Aware Importance Sampling

2024 · arXiv (Cornell University)

This paper improves upon existing data pruning methods for image classification by introducing a novel pruning metric and pruning procedure based on importance sampling. The proposed pruning metric explicitly accounts for data separability, data integrity, …

Rethinking Tamper-Evident Logging: A High-Performance, Co-Designed Auditing System

2025

Existing tamper-evident logging systems suffer from high overhead and severe data loss in high-load settings, yet only provide coarse-grained tamper detection. Moreover, installing such systems requires recompiling kernel code. To address these challenges, we present …

PLaST: Towards Paralinguistic-aware Speech Translation

2026 · Proceedings of the AAAI Conference on Artificial Intelligence

Speech translation (ST) aims to translate speech from a source language into text in the target language. Naturally, speech signals contain paralinguistic cues beyond linguistic content, which could influence or even alter the interpretation of …

QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehension

2018 · arXiv (Cornell University)

Current end-to-end machine reading and question answering (Q\&A) models are primarily based on recurrent neural networks (RNNs) with attention. Despite their success, these models are often slow for both training and inference due to the …