Yifeng Gao
YOU?
Author Swipe
View article: Self Speculative Decoding for Diffusion Large Language Models
Self Speculative Decoding for Diffusion Large Language Models Open
Diffusion-based Large Language Models (dLLMs) have emerged as a competitive alternative to autoregressive models, offering unique advantages through bidirectional attention and parallel generation paradigms. However, the generation results…
View article: Polynomial Closed Form Model for Ultra-Wideband Transmission Systems
Polynomial Closed Form Model for Ultra-Wideband Transmission Systems Open
Ultrafast and accurate physical layer models are essential for designing, optimizing and managing ultra-wideband optical transmission systems. We present a closed-form GN/EGN model, named Polynomial Closed-Form Model (PCFM), improving reli…
View article: Discrete Element Method (DEM) Studies on Correcting the Particle Size Effect on the Shear Behaviors of Gravelly Soils
Discrete Element Method (DEM) Studies on Correcting the Particle Size Effect on the Shear Behaviors of Gravelly Soils Open
The presence of overlarge gravel particles poses significant challenges for laboratory testing on prototype gravelly soils due to sample size limitations. To address this issue, replacement techniques, such as substituting overlarge partic…
View article: Token Pruning in Multimodal Large Language Models: Are We Solving the Right Problem?
Token Pruning in Multimodal Large Language Models: Are We Solving the Right Problem? Open
Multimodal large language models (MLLMs) have shown remarkable performance for cross-modal understanding and generation, yet still suffer from severe inference costs. Recently, abundant works have been proposed to solve this problem with t…
View article: Efficient Hierarchical Contrastive Self-supervising Learning for Time Series Classification via Importance-aware Resolution Selection
Efficient Hierarchical Contrastive Self-supervising Learning for Time Series Classification via Importance-aware Resolution Selection Open
Recently, there has been a significant advancement in designing Self-Supervised Learning (SSL) frameworks for time series data to reduce the dependency on data labels. Among these works, hierarchical contrastive learning-based SSL framewor…
View article: Token Pruning in Multimodal Large Language Models: Are We Solving the Right Problem?
Token Pruning in Multimodal Large Language Models: Are We Solving the Right Problem? Open
View article: Entropy-Infused Deep Learning Loss Function for Capturing Extreme Values in Wind Power Forecasting
Entropy-Infused Deep Learning Loss Function for Capturing Extreme Values in Wind Power Forecasting Open
Extreme scenarios in wind power generation occur with higher frequency and larger magnitude in the recent years due to the ever-increasing extreme meteorological factors. Accurate forecasting of the occurrence of extreme values in wind pow…
View article: Determination of elastic loss of piezoelectric materials by impedance curve fitting using intelligent algorithms
Determination of elastic loss of piezoelectric materials by impedance curve fitting using intelligent algorithms Open
Understanding the loss parameters of piezoelectric materials is crucial for designing effective piezoelectric sensors. Traditional elastic loss parameter measurement techniques mainly rely on three methods: 3 dB bandwidth, impedance fittin…
View article: Efficient High-Resolution Time Series Classification via Attention Kronecker Decomposition
Efficient High-Resolution Time Series Classification via Attention Kronecker Decomposition Open
The high-resolution time series classification problem is essential due to the increasing availability of detailed temporal data in various domains. To tackle this challenge effectively, it is imperative that the state-of-the-art attention…
View article: PMP: Privacy-Aware Matrix Profile against Sensitive Pattern Inference for Time Series
PMP: Privacy-Aware Matrix Profile against Sensitive Pattern Inference for Time Series Open
Recent rapid development of sensor technology has allowed massive fine-grained time series (TS) data to be collected and set the foundation for the development of data-driven services and applications. During the process, data sharing is o…
View article: Robust Time Series Chain Discovery with Incremental Nearest Neighbors
Robust Time Series Chain Discovery with Incremental Nearest Neighbors Open
Time series motif discovery has been a fundamental task to identify meaningful repeated patterns in time series. Recently, time series chains were introduced as an expansion of time series motifs to identify the continuous evolving pattern…
View article: Response of soil moisture to vegetation and trade-off analysis in the hilly area of the Loess Plateau, China
Response of soil moisture to vegetation and trade-off analysis in the hilly area of the Loess Plateau, China Open
It is said that in a synergistic relationship between vegetation and soil moisture (SM), the latter may be consumed excessively, while at the same time improving the ecological environment, such as in the large-scale artificial vegetation …
View article: Federated Learning with Erroneous Communication Links
Federated Learning with Erroneous Communication Links Open
In this paper, we consider the federated learning (FL) problem in the presence of communication errors. We model the link between the devices and the central node (CN) by a packet erasure channel, where the local parameters from devices ar…
View article: Differential Privacy in Privacy-Preserving Big Data and Learning: Challenge and Opportunity
Differential Privacy in Privacy-Preserving Big Data and Learning: Challenge and Opportunity Open
Differential privacy (DP) has become the de facto standard of privacy preservation due to its strong protection and sound mathematical foundation, which is widely adopted in different applications such as big data analysis, graph data proc…
View article: Towards Accurate Run-Time Hardware-Assisted Stealthy Malware Detection: A Lightweight, yet Effective Time Series CNN-Based Approach
Towards Accurate Run-Time Hardware-Assisted Stealthy Malware Detection: A Lightweight, yet Effective Time Series CNN-Based Approach Open
According to recent security analysis reports, malicious software (a.k.a. malware) is rising at an alarming rate in numbers, complexity, and harmful purposes to compromise the security of modern computer systems. Recently, malware detectio…
View article: Development of Novel Approaches to Anomaly Detection and Surety for Safeguards Data - Year One Results.
Development of Novel Approaches to Anomaly Detection and Surety for Safeguards Data - Year One Results. Open
View article: Towards superior fatigue crack growth resistance of TC4-DT alloy by in-situ rolled wire-arc additive manufacturing
Towards superior fatigue crack growth resistance of TC4-DT alloy by in-situ rolled wire-arc additive manufacturing Open
Titanium alloys have many advanced applications where they are subject to fatigue. Here, we compare the fatigue crack growth resistance and microstructures of TC4-DT alloys fabricated by wire-arc additive manufacturing (WAAM; sample S1) an…
View article: Development of Novel Approaches to Anomaly Detection and Surety for Safeguards Data - Year Two and Three Results.
Development of Novel Approaches to Anomaly Detection and Surety for Safeguards Data - Year Two and Three Results. Open
View article: cPCN-Regulated SnO2 Composites Enables Perovskite Solar Cell with Efficiency Beyond 23%
cPCN-Regulated SnO2 Composites Enables Perovskite Solar Cell with Efficiency Beyond 23% Open
Efficient electron transport layers (ETLs) not only play a crucial role in promoting carrier separation and electron extraction in perovskite solar cells (PSCs) but also significantly affect the process of nucleation and growth of the pero…
View article: MMAP: a cloud computing platform for mining the maximum accuracy of predicting phenotypes from genotypes
MMAP: a cloud computing platform for mining the maximum accuracy of predicting phenotypes from genotypes Open
Accurately predicting phenotypes from genotypes holds great promise to improve health management in humans and animals, and breeding efficiency in animals and plants. Although many prediction methods have been developed, the optimal method…
View article: TapNet: Multivariate Time Series Classification with Attentional Prototypical Network
TapNet: Multivariate Time Series Classification with Attentional Prototypical Network Open
With the advance of sensor technologies, the Multivariate Time Series classification (MTSC) problem, perhaps one of the most essential problems in the time series data mining domain, has continuously received a significant amount of attent…
View article: Semantic Discord: Finding Unusual Local Patterns for Time Series
Semantic Discord: Finding Unusual Local Patterns for Time Series Open
Finding anomalous subsequence in a long time series is a very important but difficult problem. Existing state-of-the-art methods have been focusing on searching for the subsequence that is the most dissimilar to the rest of the subsequence…
View article: Ensemble Grammar Induction For Detecting Anomalies in Time Series
Ensemble Grammar Induction For Detecting Anomalies in Time Series Open
Time series anomaly detection is an important task, with applications in a broad variety of domains. Many approaches have been proposed in recent years, but often they require that the length of the anomalies be known in advance and provid…
View article: Semantic Discord: Finding Unusual Local Patterns for Time Series
Semantic Discord: Finding Unusual Local Patterns for Time Series Open
Finding anomalous subsequence in a long time series is a very important but difficult problem. Existing state-of-the-art methods have been focusing on searching for the subsequence that is the most dissimilar to the rest of the subsequence…
View article: Ensemble Grammar Induction For Detecting Anomalies in Time Series
Ensemble Grammar Induction For Detecting Anomalies in Time Series Open
Time series anomaly detection is an important task, with applications in a broad variety of domains. Many approaches have been proposed in recent years, but often they require that the length of the anomalies be known in advance and provid…
View article: Discovering Subdimensional Motifs of Different Lengths in Large-Scale Multivariate Time Series
Discovering Subdimensional Motifs of Different Lengths in Large-Scale Multivariate Time Series Open
Detecting repeating patterns of different lengths in time series, also called variable-length motifs, has received a great amount of attention by researchers and practitioners. Despite the significant progress that has been made in recent …
View article: Morning surface temperature inversions (MSTIS) from Allegheny County,PA to Beijing, China: formation factors, health effects, and applications
Morning surface temperature inversions (MSTIS) from Allegheny County,PA to Beijing, China: formation factors, health effects, and applications Open
In recent years, concern about air quality has increased as we better understand the relationship between air pollution and health, not only of humans and animals but also of the environment. This essay looks at the ways that scientists ca…
View article: Accelerating-particle-deposition method for quickly evaluating long-term performance of fin-and-tube heat exchangers.
Accelerating-particle-deposition method for quickly evaluating long-term performance of fin-and-tube heat exchangers. Open
Fin-and-tube heat exchanger is the most commonly used heat exchanger type in air-conditioning systems. In the actual operation of air-conditioning systems, the dust particles involved in the air may partly deposit and form particulate foul…
View article: Efficient Discovery of Variable-length Time Series Motifs with Large Length Range in Million Scale Time Series
Efficient Discovery of Variable-length Time Series Motifs with Large Length Range in Million Scale Time Series Open
Detecting repeated variable-length patterns, also called variable-length motifs, has received a great amount of attention in recent years. Current state-of-the-art algorithm utilizes fixed-length motif discovery algorithm as a subroutine t…
View article: A new structural optimization method for distributors in R290 air conditioner with small diameter copper tubes.
A new structural optimization method for distributors in R290 air conditioner with small diameter copper tubes. Open
R290 (propane) is one of the most potential natural working fluids due to its zero Ozone Depletion Potential (ODP) and low Global Warming Potential (GWP). However, the apply of R290 in air conditioner is inevitably limited because of its i…