Author name cluster

Ying Luo

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

3 papers

1 author row

AAAI Conference 2026 Conference Paper

Modulation-Based Backdoors: Leveraging Amplitude and Frequency Patterns to Attack Speaker Recognition

Hanbo Cai
Pengcheng Zhang
Yan Xiao
De Li
Hanting Chu
Ying Luo

Deep neural networks (DNNs) are widely and successfully applied in the field of speaker recognition. However, recent studies reveal that these models are vulnerable to backdoor attacks, where adversaries inject malicious behaviors into victim models by poisoning the training process. Existing attack methods often rely on environmental noise or complex voice transformations, which are typically difficult to implement and exhibit poor stealthiness. To address these issues, this paper proposes two modulation-based backdoor attacks that leverage frequency modulation (FM) and amplitude modulation (AM) to construct audio triggers. In real-world scenarios, regular variations in frequency and amplitude are often imperceptible to human listeners, making the proposed attacks more covert. Experimental results show that our methods achieve high attack success rates in both digital and physical settings, while also demonstrating strong resistance to various state-of-the-art backdoor defenses.

PDF Details DOI

EAAI Journal 2023 Journal Article

A non-stationary channel prediction method for UAV communication network with error compensation

Qiuyun Zhang
Tingting Yang
Chun Wu
FanRong Shi
Hong Jiang
Qiumei Guo
Liping Deng
Ying Luo

In an unmanned aerial vehicle (UAV) communication network, especially for mission-critical applications, ultra-reliable and low-latency communication (URLLC) of the control links has essential implications for realizing collision avoidance and real-time control of UAVs. However, the UAVs’ mobility and the channel environment’s variability may make the wireless channels highly non-stationary. Real-time and accurate Channel Status Information (CSI) acquisition is a critical challenge to implementing URLLC for UAV control links. Predicting the characteristics of non-stationary channels is profitable for formulating communication strategies to mitigate the effects of future channel fading. Nevertheless, the difficulty lies in extracting accurate prediction models for non-stationary channels to obtain accurate CSI. This paper proposes a more precise prediction by utilizing the CSI obtained as feedback from the receiver and introducing error correction. The method contains three stages, CSI collection and processing, channel tracking, and error correction. First, the transmitter collects the CSI feedback from the receiver and converts it to a stationary series. Secondly, the recent historical CSIs are used to track the time evolution of the channel based on an autoregressive integrated moving average (ARIMA) model. Next, the Gaussian process regression (GPR) model is used to establish the prediction error regression model and obtain a more accurate prediction. Finally, the effectiveness of the proposed method is verified based on a UAV wireless communication network application scenario in an urban environment. Simulation results show that the proposed method outperforms other methods regarding RMSE and reliability in high-dynamic UAV communication scenarios. Significantly when the channel changes rapidly, this method can respond to the changes faster and predict the CSI more accurately.

Details DOI

AAAI Conference 2020 Conference Paper

Hierarchical Contextualized Representation for Named Entity Recognition

Ying Luo
Fengshun Xiao
Hai Zhao

Named entity recognition (NER) models are typically based on the architecture of Bi-directional LSTM (BiLSTM). The constraints of sequential nature and the modeling of single input prevent the full utilization of global information from larger scope, not only in the entire sentence, but also in the entire document (dataset). In this paper, we address these two deﬁciencies and propose a model augmented with hierarchical contextualized representation: sentence-level representation and document-level representation. In sentencelevel, we take different contributions of words in a single sentence into consideration to enhance the sentence representation learned from an independent BiLSTM via label embedding attention mechanism. In document-level, the key-value memory network is adopted to record the document-aware information for each unique word which is sensitive to similarity of context information. Our two-level hierarchical contextualized representations are fused with each input token embedding and corresponding hidden state of BiLSTM, respectively. The experimental results on three benchmark NER datasets (CoNLL-2003 and Ontonotes 5. 0 English datasets, CoNLL-2002 Spanish dataset) show that we establish new state-of-the-art results.

PDF Details