Author name cluster

Yi 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.

11 papers

2 author rows

EAAI Journal 2026 Journal Article

Credible early fire detection via intelligent pan–tilt-zoom camera

Tian Deng
Pengpai Qiu
Xiaole Lv
Edmore Tarambiwa
Yi Luo
Wenbo Xu
Ang Bian
Andreas Nienkötter

Fires cause massive losses of human life and property due to their suddenness, rapid spread, and difficult containment. Therefore, early fire detection in large-scale areas is crucial for disaster mitigation. Despite advances in object detection, identifying incipient fires using optical detectors remains hindered by minuscule flames and semi-transparent smoke. To address this, we propose a two-stage early fire detection strategy combined with optical zoom technology. We construct the You Only Look Once with cross-spatial local attention (YOLO-CLA) network. It incorporates a novel cross-spatial local attention (CLA) mechanism and an efficient partial self-attention aggregation (EPSAA) module for small target detection, alongside a lightweight dual-detection (LDDetect) head to boost computational efficiency. A filtering strategy then identifies early fire candidates based on confidence and bounding box size. Subsequently, a pan–tilt-zoom (PTZ) camera-based automatic tracking algorithm adjusts the angle and focal length to capture magnified regions of interest, which YOLO-CLA reexamines to provide reliable early-fire alarms. We also introduce a fire hazard image dataset featuring a high-resolution (HR) subset simulating small flames in wide-field monitoring. Experimental results show our model achieves 92. 41% mean Average Precision at 50% Intersection over Union threshold ( mAP 50 ) and 77. 72% mean Average Precision at Intersection over Union thresholds from 0. 50 to 0. 95 ( mAP 50: 95 ) for fire and smoke detection. It outperforms mainstream You Only Look Once (YOLO) models with 7. 63% fewer parameters and 15. 27% fewer floating-point operations (FLOPs). For small flames, our framework boosts accuracy from 42. 60% to 82. 67%, ensuring reliable detection in large-scale, long-distance scenarios.

Details DOI

EAAI Journal 2026 Journal Article

Hard constraints and soft learning dual-graph anomaly detection for industrial processes

Chuan Zhang
Ming-Qing Zhang
Yi Luo
Wei Ke
Qun-Xiong Zhu
Yan-Lin He
Yang Zhang
Yuan Xu

Anomaly detection is critical for safe and stable operation in industrial processes. Industrial data exhibits strong spatiotemporal dependence, while variable interactions often evolve dynamically. Traditional methods struggle to model both fixed physical constraints and dynamic data relationships. This paper proposes a hard constraints and soft learning dual-graph anomaly detection (HCSL-DGAD). First, a macro-graph with hard constraints is constructed based on the connections between various components of the industrial process, and micro-graphs with soft learning is constructed through an adaptive method based on attention mechanisms. The macro-graph transmits physical constraints through a spatiotemporal graph convolutional network to ensure the rationality of the abnormal propagation path. The micro-graph uses a multi-head attention mechanism to dynamically learn the implicit relationship between nodes and capture coupling information not covered by the physical topology. Secondly, to address the multi-scale anomalies in the spatiotemporal domain, a dual-channel architecture is employed to extract features from both the macro and micro-graphs. Edge weights and node states are alternately updated in the micro-graph channel to accurately identify anomaly patterns at different scales. At the same time, temporal attention and variable attention are combined in the macro-graph channel to jointly improve the detection accuracy. Extensive experiments on three benchmarks show that HCSL-DGAD achieves average F1-scores of 98. 24%, 89. 92%, and 87. 00% on the Tennessee Eastman (TE) process, Secure Water Treatment (SWaT) and PROcess NeTwork Optimization (PRONTO) datasets, respectively.

Details DOI

AAAI Conference 2026 Conference Paper

ShieldRAG: Safeguarding Retrieval-Augmented Generation from Untrusted Knowledge Bases

Peiru Yang
Haoran Zheng
Yi Luo
Xinyi Liu
Jinrui Wang
Huili Wang
Xintian Li
Yongfeng Huang

Open knowledge bases (e.g., websites) are widely adopted in Retrieval-Augmented Generation (RAG) systems to provide supplementary knowledge (e.g., latest information). However, such sources inevitably contain biased or harmful content, and incorporating these untrusted contents into the RAG process introduces significant safety risks, including the degradation of LLM performance and the potential generation of harmful outputs. Recent studies have shown that this vulnerability can be further amplified by adversarial poisoning attacks specifically targeting the knowledge sources. Most existing methods primarily emphasize improving the accuracy and efficiency of RAG systems, usually overlooking these critical safety concerns. In this paper, we propose a safety-aware retrieval framework (ShieldRAG) designed to augment language model generation by jointly optimizing for both relevance and safety in the retrieved knowledge content. The core idea of ShieldRAG is to transfer the safety knowledge implicitly encoded in powerful LLMs into the retriever model through an adversarial knowledge alignment mechanism. This can empower the retriever with the safety awareness, and adapt to the diverse and unknown distribution of unsafe content encountered in practical scenarios. We evaluate ShieldRAG on seven real-world datasets using five widely-used LLMs and two state-of-the-art poisoning attack strategies. Experimental results show that our method substantially improves the robustness of RAG systems against unsafe knowledge sources, while maintaining competitive performance in terms of generation accuracy and efficiency.

PDF Details DOI

EAAI Journal 2025 Journal Article

Latent temporal smoothness-induced Schatten- p norm factorization for sequential subspace clustering

Yuan Xu
Zhen-Zhen Zhao
Tong-Wei Lu
Wei Ke
Yi Luo
Yan-Lin He
Qun-Xiong Zhu
Yang Zhang

This paper presents an innovative latent temporal smoothness-induced Schatten- p norm factorization (SpFLTS) method aimed at addressing challenges in sequential subspace clustering tasks. Globally, SpFLTS employs a low-rank subspace clustering framework based on Schatten-2/3 norm factorization to enhance the comprehensive capture of the original data features. Locally, a total variation smoothing term is induced to the temporal gradients of latent subspace matrices obtained from sub-orthogonal projections, thereby preserving smoothness in the sequential latent space. To efficiently solve the closed-form optimization problem, a fast Fourier transform is combined with the non-convex alternating direction method of multipliers to optimize latent subspace matrix, which greatly speeds up computation. Experimental results demonstrate that the proposed SpFLTS method surpasses existing techniques on multiple benchmark databases, highlighting its superior clustering performance and extensive application potential.

Details DOI

AAAI Conference 2024 Conference Paper

SECap: Speech Emotion Captioning with Large Language Model

Yaoxun Xu
Hangting Chen
Jianwei Yu
Qiaochu Huang
Zhiyong Wu
Shi-Xiong Zhang
Guangzhi Li
Yi Luo

Speech emotions are crucial in human communication and are extensively used in fields like speech synthesis and natural language understanding. Most prior studies, such as speech emotion recognition, have categorized speech emotions into a fixed set of classes. Yet, emotions expressed in human speech are often complex, and categorizing them into predefined groups can be insufficient to adequately represent speech emotions. On the contrary, describing speech emotions directly by means of natural language may be a more effective approach. Regrettably, there are not many studies available that have focused on this direction. Therefore, this paper proposes a speech emotion captioning framework named SECap, aiming at effectively describing speech emotions using natural language. Owing to the impressive capabilities of large language models in language comprehension and text generation, SECap employs LLaMA as the text decoder to allow the production of coherent speech emotion captions. In addition, SECap leverages HuBERT as the audio encoder to extract general speech features and Q-Former as the Bridge-Net to provide LLaMA with emotion-related speech features. To accomplish this, Q-Former utilizes mutual information learning to disentangle emotion-related speech features and speech contents, while implementing contrastive learning to extract more emotion-related speech features. The results of objective and subjective evaluations demonstrate that: 1) the SECap framework outperforms the HTSAT-BART baseline in all objective evaluations; 2) SECap can generate high-quality speech emotion captions that attain performance on par with human annotators in subjective mean opinion score tests.

PDF Details DOI

YNIMG Journal 2024 Journal Article

The acquired dyad inclination and decreased interpersonal brain communication in the pursuit of collective benefit

Shuyi Li
Linwei Yu
Xiaorong Gan
Yingying Hou
Yafeng Pan
Yi Luo
Yi Hu

People perform better collectively than individually, a phenomenon known as the collective benefit. To pursue the benefit, they may learn from previous behaviors, come to know whose initial opinion should be valued, and develop the inclination to take it as the collective one. Such learning may affect interpersonal brain communication. To test these hypotheses, this study recruited participant dyads to conduct a perceptual task on which they made individual decisions first and then the collective one. The enhanced interpersonal brain synchronization (IBS) between participants was explored when individual decisions were in disagreement vs. agreement. Computational modeling revealed that participant dyads developed the dyad inclination of taking the higher-able participants', not the lower-able ones' decisions as their collective ones. Brain analyses unveiled the enhanced IBS at frontopolar areas, premotor areas, supramarginal gyri, and right temporal-parietal junctions. The premotor IBS correlated negatively with dyad inclination and collective benefit in the absence of correction. The Granger causality analyses further supported the negative relation of dyad inclination with inter-brain communication. This study highlights that dyads learn to weigh individuals' decisions, resulting in dyad inclinations, and explores associated inter-brain communication, offering insights into the dynamics of collective decision-making.

Details DOI

YNIMG Journal 2021 Journal Article

From gratitude to injustice: Neurocomputational mechanisms of gratitude-induced injustice

Ruida Zhu
Zhenhua Xu
Song Su
Chunliang Feng
Yi Luo
Honghong Tang
Shen Zhang
Xiaoyan Wu

Gratitude shapes individuals' behaviours and impacts the harmony of society. Many previous studies focused on its association with prosocial behaviours. A possibility that gratitude can lead to moral violation has been overlooked until recently. Nevertheless, the neurocognitive mechanisms of gratitude-induced moral violation are still unclear. On the other hand, though neural correlates of the gratitude's formation have been examined, the neural underpinnings of gratitude-induced behaviour remain unknown. For addressing these two overlapped research gaps, we developed novel tasks to investigate how participants who had received voluntary (Gratitude group) or involuntary help (Control group) punished their benefactors' unfairness with functional magnetic resonance imaging (fMRI). The Gratitude group punished their benefactors less than the Control group. The self-report and computational modelling results demonstrated a crucial role of the boosted protection tendency on behalf of benefactors in the gratitude-induced injustice. The fMRI results showed that activities in the regions associated with mentalizing (temporoparietal junction) and reward processing (ventral medial prefrontal cortex) differed between the groups and were related to the gratitude-induced injustice. They suggest that grateful individuals concern for benefactors' benefits, value chances to interact with benefactors, and refrain from action that perturbs relationship-building (i.e., exert less punishment on benefactors' unfairness), which reveal a dark side of gratitude and enrich the gratitude theory (i.e., the find-bind-remind theory). Our findings provide psychological, computational, and neural accounts of the gratitude-induced behaviour and further the understanding of the nature of gratitude.

Details DOI

NeurIPS Conference 2021 Conference Paper

Speech Separation Using an Asynchronous Fully Recurrent Convolutional Neural Network

Xiaolin Hu
Kai Li
Weiyi Zhang
Yi Luo
Jean-Marie Lemercier
Timo Gerkmann

Recent advances in the design of neural network architectures, in particular those specialized in modeling sequences, have provided significant improvements in speech separation performance. In this work, we propose to use a bio-inspired architecture called Fully Recurrent Convolutional Neural Network (FRCNN) to solve the separation task. This model contains bottom-up, top-down and lateral connections to fuse information processed at various time-scales represented by stages. In contrast to the traditional approach updating stages in parallel, we propose to first update the stages one by one in the bottom-up direction, then fuse information from adjacent stages simultaneously and finally fuse information from all stages to the bottom stage together. Experiments showed that this asynchronous updating scheme achieved significantly better results with much fewer parameters than the traditional synchronous updating scheme on speech separation. In addition, the proposed model achieved competitive or better results with high efficiency as compared to other state-of-the-art approaches on two benchmark datasets.

PDF Details

AAAI Conference 2018 Conference Paper

Multi-Task Medical Concept Normalization Using Multi-View Convolutional Neural Network

Yi Luo
Guojie Song
Pengyu Li
Zhongang Qi

Medical concept normalization is a critical problem in biomedical research and clinical applications. In this paper, we focus on normalizing diagnostic and procedure names in Chinese discharge summaries to standard entities, which is formulated as a semantic matching problem. However, non-standard Chinese expressions, short-text normalization and heterogeneity of tasks pose critical challenges in our problem. This paper presents a general framework which introduces a tensor generator and a novel multi-view convolutional neural network (CNN) with multi-task shared structure to tackle the two tasks simultaneously. We propose that the key to address non-standard expressions and short-text problem is to incorporate a matching tensor with multiple granularities. Then multi-view CNN is adopted to extract semantic matching patterns and learn to synthesize them from different views. Finally, multi-task shared structure allows the model to exploit medical correlations between disease and procedure names to better perform disambiguation tasks. Comprehensive experimental analysis indicates our model outperforms existing baselines which demonstrates the effectiveness of our model.

PDF Details

AAMAS Conference 2007 Conference Paper

Children in the Forest: Towards a Canonical Problem of Spatio-Temporal Collaboration

Yi Luo
Ladislau Bölöni

Canonical problems are simplified representations of a class of real world problems. They allow researchers to compare algorithms in a standard setting which captures the most important challenges of the real world problems being modeled. Such examples are the block world for planning, two-player games for algorithms which learn the behavior of the opponent agent, or the "split the pie" game for a large class of negotiation problems.

PDF

IROS Conference 2006 Conference Paper

Vision-Based Assembly of Capillary for Microfluidic Device

Xiaodong Wang 0021
Xiujun Wang
Yi Luo
Chong Liu
Liqun Ma

The application of plastic microfluidic chips can be extended with quartz capillaries connected at the end of their microchannels, e. g. UV absorption detection method can be carried out, which responds to almost 80% chemical compounds in detection. A vision-based experiment system for automatically assembling capillaries to plastic microfluidic chips was set up. UN-curing adhesive is used for the joining procedure. Visual feedback is implemented in the assembly system and the control algorithm is briefly introduced. The methods for obtaining the spatial position deviation between the capillary and the end of the chip's micro channel are described. The deviation information in x-y plane is obtained by performing image processing and converting pixels into actual distance with calibrated data. Two methods for recovering vertical deviation information were explored, and both are feasible for application, one is the depth-in-focus, the other is with the use of microscopic stereovision.

Details