A5100777469- Cryptographic Implementations and Security
- Physical Unclonable Functions (PUFs) and Hardware Security
- Antenna Design and Analysis
- Chaos-based Image/Signal Encryption
- Digital Media Forensic Detection
- Natural Language Processing Techniques
- Blind Source Separation Techniques
- Wireless Signal Modulation Classification
- Topic Modeling
- Computer Graphics and Visualization Techniques
- Ultra-Wideband Communications Technology
- Electromagnetic Compatibility and Noise Suppression
- Advanced Image and Video Retrieval Techniques
- EEG and Brain-Computer Interfaces
- Advanced Antenna and Metasurface Technologies
- Microwave Engineering and Waveguides
- Optical Network Technologies
- Advanced Fiber Laser Technologies
- Power Line Communications and Noise
- Millimeter-Wave Propagation and Modeling
- 3D Shape Modeling and Analysis
- Advanced Malware Detection Techniques
- Speech and Audio Processing
- Data Visualization and Analytics
- Coding theory and cryptography
Hebei University of Technology
2025
Beijing University of Posts and Telecommunications
2015-2024
Zhejiang University
2014-2024
Northeast Agricultural University
2024
Institute of Automation
2023
Chinese Academy of Sciences
2009-2023
University of Chinese Academy of Sciences
2023
Stanford University
2023
Zhoukou Normal University
2021-2023
Shanghai Jiao Tong University
2022-2023
Recent AI-assistant agents, such as ChatGPT, predominantly rely on supervised fine-tuning (SFT) with human annotations and reinforcement learning from feedback (RLHF) to align the output of large language models (LLMs) intentions, ensuring they are helpful, ethical, reliable. However, this dependence can significantly constrain true potential agents due high cost obtaining supervision related issues quality, reliability, diversity, self-consistency, undesirable biases. To address these...
Generating a Chain of Thought (CoT) has been shown to consistently improve large language model (LLM) performance on wide range NLP tasks. However, prior work mainly focused logical reasoning tasks (e.g. arithmetic, commonsense QA); it remains unclear whether improvements hold for more diverse types reasoning, especially in socially situated contexts. Concretely, we perform controlled evaluation zero-shot CoT across two sensitive domains: harmful questions and stereotype benchmarks. We find...
In this paper, we propose a novel shape interpolation approach based on Poisson equation. We formulate the trajectory problem of as solving equations defined domain mesh. A non-linear gradient field method is proposed to take both vertex coordinates and surface orientation into account. With proper boundary conditions, in-between shapes are reconstructed implicitly from interpolated fields, while traditional methods usually manipulate directly. Besides global interpolation, our also...
Wire arc additive manufacturing (WAAM) is a viable technology for complex and medium-to-large-sized invar alloy components. However, the cyclic thermal input during WAAM process may cause austenite grains in component to grow abnormally, adversely impacting material’s mechanical properties. The addition of alloying elements such as Cr, Mo, V can refine microstructure solve these problems. This study examines influence V, N on properties produced through wire manufacturing. form various...
Abstract This article is focused on the electromagnetic radiation effects of Silicon Carbide (SiC) Metal-Oxide-Semiconductor Field-Effect Transistors (MOSFETs) under total ionizing dose (TID) effect. A 650V trench-gate SiC MOSFET was selected as experimental subject. The variations in electronic characteristics MOSFETs were explained from perspective crystal defects. Furthermore, an analysis trends device conducted based changes electrical properties devices. broad network model for signals...
Abstract Background Previous studies have revealed that circadian genes play important roles in cell proliferation, apoptosis, cycle control, DNA damage response and treatment of chemotherapy agents cancers. Aims We hypothesized the polymorphisms may be associated with prognosis hepatocellular carcinoma ( HCC ) patients treated transcatheter arterial chemoembolization TACE ). Methods Twelve functional single nucleotide SNP s) negative feedback regulation (including CRY 1, 2, PER 2 3 were...
Cloud computing is an essential technology to Big Data analytics and services. A cloud system often comprised of a large number parallel storage devices. Monitoring the usage performance such important for efficient operations, maintenance, security. Tracing every application on untenable due scale privacy issues. But profile data can be collected relatively efficiently by regularly sampling state system, including properties as CPU load, memory usage, network others, creating set...
The fiber optical network offers high speed, large bandwidth, and a degree of reliability. However, the development communication technology has hit bottleneck due to several challenges such as energy loss, cost, system capacity approaching Shannon limit. As powerful tool, machine learning provides strong driving force for various industries greatly promotes society. Machine also new possible solution achieve greater transmission capacities longer distances in communications. In this...
The increasing popularity of Internet Things (IoT) devices provides a huge data source for intelligent identification. Images captured by the unmanned aerial vehicle (UAV) are often exploited in target detection missions search, rescue and fire prevention. However, without sufficient training samples, machine learning method is usually difficult to meet application requirements. To solve problem, brain–computer interface (BCI) real-time system applied UAV detection. In this study, novel...
The limits of open-ended generative models are unclear, yet increasingly important. What causes them to succeed and what fail? In this paper, we take a prompt-centric approach analyzing bounding the abilities models. We present generic methodology analysis with two challenging prompt constraint types: structural stylistic. These types categorized into set well-defined constraints that analyzable by single prompt. then systematically create diverse simple, natural, useful prompts robustly...
Explicitly sharing individual level data in genomics studies has many merits comparing to summary statistics, including more strict QCs, common statistical analyses, relative identification and improved power GWAS, but it is hampered by privacy or ethical constraints. In this study, we developed encG-reg, a regression approach that can detect relatives of various degrees based on encrypted genomic data, which immune The encryption properties encG-reg are the random matrix theory masking...
Abstract For wireless transmission, radio-frequency device anti-cloning has become a major security issue. Radio-frequency distinct native attribute (RF-DNA) fingerprint is developing technology to find the difference among RF devices and identify them. Comparing with previous research, (1) this paper proposed that mean ( μ ) feature should be added into RF-DNA fingerprint. Thus, totally four statistics (mean, standard deviation, skewness, kurtosis) were calculated on instantaneous...
We propose and experimentally demonstrate single-sideband (SSB) photonic vector millimeter-wave (mm-wave) signal generation enabled by optical dual-SSB modulation a single push-pull Mach-Zehnder modulator (MZM). use software-based digital processing to generate the driving for MZM. The has vector-modulated sideband an unmodulated sideband, which are spaced our desired mm-wave carrier frequency. After linear of MZM, filtering, single-ended photodiode detection, we can get electrical...
Abstract Objective. Asynchronous brain-computer interfaces (BCIs) show significant advantages in many practical application scenarios. Compared with the rapid development of synchronous BCIs technology, progress asynchronous BCI research, terms containing multiple targets and training-free detection, is still relatively slow. In order to improve practicability BCI, a spatio-temporal equalization multi-window algorithm (STE-MW) was proposed for detection steady-state visual evoked potential...