Tao Zhang

dm-x

2017

The verified boot feature in recent Android devices, which deploys dm-verity, has been overwhelmingly successful in eliminating the extremely popular Android smart phone rooting movement [25]. Unfortunately, dm-verity integrity guarantees are read-only and do not …

A Formal Verification Method of Obligation Policy in Multi-agent System

2015 · International Journal of u- and e- Service Science and Technology

In Multi-Agent System, obligations are actions that agents are required to take or some states of affairs which should be maintained, formal modeling and verifying obligation policy which is high-level requirements specifications or communication protocol …

Some new results on permutation polynomials over finite fields

2015 · arXiv (Cornell University)

Permutation polynomials over finite fields constitute an active research area and have applications in many areas of science and engineering. In this paper, four classes of monomial complete permutation polynomials and one class of trinomial …

Multi-class Unbalanced Data Classification for Sleep Staging

2020 · International Journal of Computer and Electrical Engineering

Unbalanced data classification is a research focus for many applications, including financial fraud detection, network intrusion detection and cancer classification.However, unbalanced data classification is rarely investigated in the field of EEG-based sleep staging.Herein, considering the …

Task Scheduling Algorithm of Cyber-Physical System Base on Complex Industry and Big Data

2021

In order to address the problem of unbalanced load of cyber-physical systems caused by high data concurrency, heterogeneous resources, and complex task nature in complex industrial and big data environments, this paper proposes a dynamic …

Dual-Target Cross-Domain Bundle Recommendation

2021

The conventional Cross-Domain Recommendation(CDR) approaches are single-target that focus only on improving the recommendation performance of the target domain. To enhance the performance of both source and target domains, dual-target CDR approaches have been proposed. …

Distant Supervision Relation Extraction via Reinforcement Learning with Potential Energy Function

2021

Distant supervision has become an essential method for relation extraction. Although distant supervision is very effective, there is a large amount of noise in the dataset produced by distant supervision. To filter out the noise …

Scale-Invariant Adversarial Attack for Evaluating and Enhancing Adversarial Defenses

2022 · arXiv (Cornell University)

Efficient and effective attacks are crucial for reliable evaluation of defenses, and also for developing robust models. Projected Gradient Descent (PGD) attack has been demonstrated to be one of the most successful adversarial attacks. However, …

DG-Trans: Automatic Code Summarization via Dynamic Graph Attention-based Transformer

2021 · 2021 IEEE 21st International Conference on Software Quality, Reliability and Security (QRS)

Automatic code summarization is an important topic in the software engineering field, which aims to automatically generate the description for the source code. Based on Graph Neural Networks (GNN), most existing methods apply them to …

Kernel Relative-prototype Spectral Filtering for Few-shot Learning

2022 · arXiv (Cornell University)

Few-shot learning performs classification tasks and regression tasks on scarce samples. As one of the most representative few-shot learning models, Prototypical Network represents each class as sample average, or a prototype, and measures the similarity …

Federated Learning with Server Learning: Enhancing Performance for Non-IID Data

2022 · arXiv (Cornell University)

Federated Learning (FL) has emerged as a means of distributed learning using local data stored at clients with a coordinating server. Recent studies showed that FL can suffer from poor performance and slower convergence when …

CoF-CoT: Enhancing Large Language Models with Coarse-to-Fine Chain-of-Thought Prompting for Multi-domain NLU Tasks

2023

While Chain-of-Thought prompting is popular in reasoning tasks, its application to Large Language Models (LLMs) in Natural Language Understanding (NLU) is under-explored. Motivated by multi-step reasoning of LLMs, we propose Coarse-to-Fine Chain-of-Thought (CoF-CoT) approach that …

Cost-Sensitive Hypergraph Learning With Structure Quality Preservation for IoT Software Defect Prediction

2024 · IEEE Open Journal of the Communications Society

Generative AI is revolutionizing Software Engineering (SE), as both engineers and academics embrace this technology in their work. To better leverage this technology for software generation, it is essential to propose effective IoT software defect …

An Efficient Graph Autoencoder with Lightweight Desmoothing Decoder and Long-Range Modeling

2024

Graph self-supervised learning provides a powerful guarantee for learning high-quality representations in an unsupervised manner. Despite its early birth, the performance of generative graph self-supervised learning has long lagged behind that of up-and-coming contrastive learning, …

SuperRS: Multi Scenario Reciprocal-Aware Dual MoE for Unified Recommendation-Search Ranking

2025

In e-commerce, search and recommendation rankings require a deep understanding of user behaviors and personalized scoring of products. While existing systems maintain separate pipelines for search and recommendation, these two scenarios share aligned objectives and …

Residual-PAC Privacy: Automatic Privacy Control Beyond the Gaussian Barrier

2025 · arXiv (Cornell University)

The Probably Approximately Correct (PAC) Privacy framework [46] provides a powerful instance-based methodology to preserve privacy in complex data-driven systems. Existing PAC Privacy algorithms (we call them Auto-PAC) rely on a Gaussian mutual information upper …