Author name cluster

Rong Li

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.

10 papers

1 author row

JBHI Journal 2026 Journal Article

Subject-Adaptive EEG Decoding via Filter-Bank Neural Architecture Search for BCI Applications

Chong Wang
Li Yang
Bingfan Yuan
Jiafan Zhang
Chen Jin
Rong Li
Junjie Bu

Individual differences pose a significant challenge in brain-computer interface (BCI) research. Designing a universally applicable network architecture is impractical due to the variability in human brain structure and function. We propose Filter-Bank Neural Architecture Search (FBNAS), an EEG decoding framework that automates network architecture design for individuals. FBNAS uses three temporal cells to process different frequency EEG signals, with dilated convolution kernels in their search spaces. A multi-path NAS algorithm determines optimal architectures for multi-scale feature extraction. We benchmarked FBNAS on three EEG datasets across two BCI paradigms, comparing it to six state-of-the-art deep learning algorithms. FBNAS achieved cross-session decoding accuracies of 79. 78%, 70. 66%, and 68. 38% on the BCIC-IV-2a, OpenBMI, and SEED datasets, respectively, outperforming other methods. Our results show that FBNAS customizes decoding models to address individual differences, enhancing decoding performance and shifting model design from expert-driven to machine-aided. The source code can be found at https://github.com/wang1239435478/FBNAS-master.

Details DOI

NeurIPS Conference 2025 Conference Paper

3EED: Ground Everything Everywhere in 3D

Rong Li
Yuhao Dong
Tianshuai Hu
Alan Liang
Youquan Liu
Dongyue Lu
Liang Pan
Lingdong Kong

Visual grounding in 3D is the key for embodied agents to localize language-referred objects in open-world environments. However, existing benchmarks are limited to indoor focus, single-platform constraints, and small scale. We introduce 3EED, a multi-platform, multi-modal 3D grounding benchmark featuring RGB and LiDAR data from vehicle, drone, and quadruped platforms. We provide over 128, 000 objects and 22, 000 validated referring expressions across diverse outdoor scenes -- 10x larger than existing datasets. We develop a scalable annotation pipeline combining vision-language model prompting with human verification to ensure high-quality spatial grounding. To support cross-platform learning, we propose platform-aware normalization and cross-modal alignment techniques, and establish benchmark protocols for in-domain and cross-platform evaluations. Our findings reveal significant performance gaps, highlighting the challenges and opportunities of generalizable 3D grounding. The 3EED dataset and benchmark toolkit are released to advance future research in language-driven 3D embodied perception.

PDF Details

YNIMG Journal 2025 Journal Article

A dynamic spatiotemporal representation framework for deciphering personal brain function

Xuyang Wang
Ting Zou
Haofei Wang
Honghao Han
Huafu Chen
Vince D. Calhoun
Rong Li

Functional magnetic resonance imaging (fMRI) opens a window on observing spontaneous activities of the human brain in vivo. However, the high complexity of fMRI signals makes brain functional representations intractable. Here, we introduce a state decomposition method to reduce this complexity and decipher individual brain functions at multiple levels. Briefly, brain dynamics are captured by temporal first-order derivatives and spatially divided into 'state sets' at each time point based on the velocity and direction of change. This approach transforms the original signals into discrete series consisting of four fundamental states, which efficiently encode individual-specific information. Subsequently, we designed a suite of state-based metrics to quantify regional activities and network interactions. Compared with conventional representations such as resting-state fluctuation amplitude and Pearson's functional connectivity, the state-based representations serve as more discriminative 'brain fingerprints' for individuals and produce reproducible spatial patterns across heterogeneous cohorts (n = 1015). Regarding functional organization, our proposed profiles extend previous representations into nonlinear domains, revealing not only the canonical default-mode dominant pattern but also patterns dominated by the attention network and basal ganglia. Moreover, we demonstrate that personal phenotypes (such as age and gender) can be decoded from regional representations with high accuracy. The equivalence between state series outperforms other existing network representations in predicting individual fluid intelligence. Overall, this framework establishes a foundation for enriching the repertoire of brain functional representations and enhancing the power of brain-phenotype modeling.

Details DOI

AAAI Conference 2025 Conference Paper

ConSense: Continually Sensing Human Activity with WiFi via Growing and Picking

Rong Li
Tao Deng
Siwei Feng
Mingjie Sun
Juncheng Jia

WiFi-based human activity recognition (HAR) holds significant application potential across various fields. To handle dynamic environments where new activities are continuously introduced, WiFi-based HAR systems must adapt by learning new concepts without forgetting previously learned ones. Furthermore, retaining knowledge from old activities by storing historical exemplar is impractical for WiFi-based HAR due to privacy concerns and limited storage capacity of edge devices. In this work, we propose ConSense, a lightweight and fast-adapted exemplar-free class incremental learning framework for WiFi-based HAR. The framework leverages the transformer architecture and involves dynamic model expansion and selective retraining to preserve previously learned knowledge while integrating new information. Specifically, during incremental sessions, small-scale trainable parameters that are trained specifically on the data of each task are added in the multi-head self-attention layer. In addition, a selective retraining strategy that dynamically adjusts the weights in multilayer perceptron based on the performance stability of neurons across tasks is used. Rather than training the entire model, the proposed strategies of dynamic model expansion and selective retraining reduce the overall computational load while balancing stability on previous tasks and plasticity on new tasks. Evaluation results on three public WiFi datasets demonstrate that ConSense not only outperforms several competitive approaches but also requires fewer parameters, highlighting its practical utility in class-incremental scenarios for HAR.

PDF Details DOI

YNICL Journal 2025 Journal Article

Distinct cortical morphometric inverse divergence changes in Parkinson’s disease correlate with transcriptional expression patterns

Ting Zou
Xuyang Wang
Xiaofei Hu
Qing Gao
Honghao Han
Huafu Chen
Rong Li

Growing evidence shows that parkinson's disease (PD) is a heterogeneous neurodegenerative disorder associated with region-specific changes in brain anatomy. However, the genetic mechanisms underlining these abnormalities are unclear. We aim to investigate PD neuroanatomical subtypes and uncover the specific brain-wide gene expression associated with morphometric abnormalities in each PD subtype. The morphometric inverse divergence (MIND) algorithm was used to quantify the morphological similarity based on multiple MRI features in 127 patients with PD and 101 healthy controls (HC). Then, heterogeneity through discriminant analysis (HYDRA) was employed to investigate the PD subtypes based on the MIND strength. Intergroup comparisons were conducted to assess MIND strength and clinical behavioral differences among PD subtypes and HC. Finally, we explored the associations between MIND network changes and gene expression in each PD subtype through partial least squares (PLS) regression, functional enrichment of PLS-weighted genes and transcriptional signature assessment of cell types. We identified two distinct subtypes of PD-related MIND changes, indicating that MIND decreased mainly in the frontal and cingulate cortices in subtype 1, and increased mainly in the occipital cortex and postcentral gyrus in subtype 2 (Bonferroni correction, p < 0.05). Both PD subtypes exhibited impaired cognitive function compared to HC, with subtype 2 showing lower Unified Parkinson's Disease Rating Scale Part III (UPDRS-III) and Hoehn and Yahr (H&Y) scores than subtype 1. Moreover, genetic commonalities analysis were identified 5 shared negative genes in the PD subtypes. Subtype 1 PLS1 genes were functionally enriched in biological processes related to synaptic function, neurodevelopment and degeneration. In addition, subtype 2 PLS1 genes showed additional involvement of metabolic pathways alongside synaptic function. Moreover, we identified MIND-related genes involved in inhibitory and excitatory neurons in subtype 1. In subtype 2, MIND-related genes were involved in astrocytes besides excitatory and inhibitory neurons. Our findings suggest two distinct neuroanatomical subtypes in PD, deepening the understanding of the heterogeneity of PD by bridging the gap between the transcriptome and neuroimaging.

Details DOI

YNICL Journal 2025 Journal Article

Medulla oblongata dominated synaptic density network degeneration in amyotrophic lateral sclerosis

Ting Zou
Manliu Hou
Honghao Han
Xuyang Wang
Huafu Chen
Yongxiang Tang
Rong Li
Shuo Hu

BACKGROUND: Amyotrophic lateral sclerosis (ALS) is a brain network disorder closely associated with synaptic loss in the upper and lower motor neurons. However, the in vivo synaptic network changes and their progressive processes remain unclear. Here, we aim to investigate the synaptic density network connectivity and the likely sequences of synaptic loss in patients with ALS. METHODS: F]SynVesT-1. The individual synaptic density similarity network was constructed for each patient by calculating the similarity between interregional synaptic density distributions. The synaptic network connectivity changes were investigated, followed by an examination of the local synaptic density in regions that showed significant network alterations. Finally, we constructed the voxel-wise and ROI-wise causal synaptic covariance network (cSCN) by applying Granger causality analysis. This allowed us to identify the sequence of synaptic loss in these brain regions. RESULTS: We observed an overall decrease in synaptic density network connectivity in ALS patients compared to controls, with the highest nodal degree in the right medulla oblongata. Specifically, the reduced connections were dominantly between the medulla oblongata and the striatum, frontal lobe, occipital lobe, as well as between the striatum and the frontal lobe, occipital lobe. Furthermore, patients with ALS displayed significantly synaptic loss in those brain regions. The cSCN analyses showed that as the disease progresses, the cortical synaptic loss sequences of ALS extend from the medulla oblongata to the regions including the striatum, frontal lobe, occipital lobe, and parietal lobe. CONCLUSIONS: These findings suggest that synaptic density network degeneration in ALS may follow a bottom-up transmission pattern, primarily involving in the medulla oblongata-striatum-neocortex network, which have the potential to capture new network-based targets for clinical therapy in the progression of ALS.

Details DOI

AAAI Conference 2025 Conference Paper

Region-aware Difference Distilling with Attribute-guided Contrastive Regularization for Change Captioning

Rong Li
Liang Li
Jiehua Zhang
Qiang Zhao
Hongkui Wang
Chenggang Yan

Change captioning aims to describe the differences between two similar images using natural language, significantly aiding in understanding and monitoring changes. This challenging task requires a fine-grained understanding of subtle changes while resisting disturbances like viewpoint shifts and illumination variations. Existing methods often rely solely on global difference features and lack comprehensive alignment of linguistic and visual information, leading to overlooking fine-grained details and generating semantic hallucinated sentences. To address these limitations, we propose the region-aware difference distilling (RDD) network with attribute-guided contrastive regularization (ACR). The RDD uses global difference features to progressively distill regional difference features using learnable vectors, allowing for more precise identification of changed regions. The ACR enhances comprehensive alignment between linguistic and visual information by formulating Nouns-to-Objects (N2O) and Verbs-to-Actions (V2A) alignment losses to regularize the regional difference features. Promising results on three datasets demonstrate that our method outperforms the state-of-the-art change captioning methods.

PDF Details DOI

AAAI Conference 2025 Conference Paper

Structure Balance and Gradient Matching-Based Signed Graph Condensation

Rong Li
Long Xu
Songbai Liu
Junkai Ji
Lingjie Li
Qiuzhen Lin
Lijia Ma

Training graph neural networks (GNNs) for graph representation has received increasing concerns due to its outstanding performance in the link prediction and node classification tasks, but it incurs much time and storage for tackling large-scale graphs. To alleviate this issue, graph condensation has been emerged to condense the large graph into a small but highly-informative graph, while achieving comparable performance of GNNs trained on the small graph and large graph. However, existing works mainly focus on the gradient or distribution matching under GNN training trajectories to condense simple link structures, while overlooking the structure matching for condensing signed graph that exists conflict links and structural balance among nodes. To bridge this gap, we propose a novel Structure Balance and Gradient Matching-Based Signed Graph Condensation (SGSGC) method for condensing signed graph with node attributes, conflict links and structural balance into informative smaller ones. Specifically, we first propose a structure-balanced matching to match the structural balance between the original and condensed signed graph, and then combine it with the gradient matching to condense signed graph for the link sign prediction task, while preserving both conflicting link structures and node attributes. Moreover, we use the feature smoothing and the graph sparsification technique to improve the robustness for the GNN training, respectively. Finally, a bi-level optimization technique is proposed to simultaneously find the optimal node attributes and conflict structure of the condensed graph. Experiments on six datasets demonstrate that SGSGC achieves excellent performance. On Epinions, 94% test accuracy of training on the original signed graph, while reducing their graph size by 99.95% - 99.99%, and there exist 2.24% – 6.26% accuracy improvements for link sign prediction compared to the state-of-the-arts.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

Talk2Event: Grounded Understanding of Dynamic Scenes from Event Cameras

Lingdong Kong
Dongyue Lu
Alan Liang
Rong Li
Yuhao Dong
Tianshuai Hu
Lai Xing Ng
Wei Ooi

Event cameras offer microsecond-level latency and robustness to motion blur, making them ideal for understanding dynamic environments. Yet, connecting these asynchronous streams to human language remains an open challenge. We introduce Talk2Event, the first large-scale benchmark for language-driven object grounding in event-based perception. Built from real-world driving data, Talk2Event provides over 30, 000 validated referring expressions, each enriched with four grounding attributes -- appearance, status, relation to viewer, and relation to other objects -- bridging spatial, temporal, and relational reasoning. To fully exploit these cues, we propose EventRefer, an attribute-aware grounding framework that dynamically fuses multi-attribute representations through a Mixture of Event-Attribute Experts (MoEE). Our method adapts to different modalities and scene dynamics, achieving consistent gains over state-of-the-art baselines in event-only, frame-only, and event-frame fusion settings. We hope our dataset and approach will establish a foundation for advancing multimodal, temporally-aware, and language-driven perception in real-world robotics and autonomy.

PDF Details

YNIMG Journal 2021 Journal Article

The pulse: transient fMRI signal increases in subcortical arousal systems during transitions in attention

Rong Li
Jun Hwan Ryu
Peter Vincent
Max Springer
Dan Kluger
Erik A. Levinsohn
Yu Chen
Huafu Chen

Studies of attention emphasize cortical circuits for salience monitoring and top-down control. However, subcortical arousal systems have a major influence on dynamic cortical state. We hypothesize that task-related increases in attention begin with a "pulse" in subcortical arousal and cortical attention networks, which are reflected indirectly through transient fMRI signals. We conducted general linear model and model-free analyses of fMRI data from two cohorts and tasks with mixed block and event-related design. 46 adolescent subjects at our center and 362 normal adults from the Human Connectome Project participated. We identified a core shared network of transient fMRI increases in subcortical arousal and cortical salience/attention networks across cohorts and tasks. Specifically, we observed a transient pulse of fMRI increases both at task block onset and with individual task events in subcortical arousal areas including midbrain tegmentum, thalamus, nucleus basalis and striatum; cortical-subcortical salience network regions including the anterior insula/claustrum and anterior cingulate cortex/supplementary motor area; in dorsal attention network regions including dorsolateral frontal cortex and inferior parietal lobule; as well as in motor regions including cerebellum, and left hemisphere hand primary motor cortex. The transient pulse of fMRI increases in subcortical and cortical arousal and attention networks was consistent across tasks and study populations, whereas sustained activity in these same networks was more variable. The function of the transient pulse in these networks is unknown. However, given its anatomical distribution, it could participate in a neuromodulatory surge of activity in multiple parallel neurotransmitter systems facilitating dynamic changes in conscious attention.

Details DOI