Author name cluster

Liwen Zhang

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.

15 papers

2 author rows

AAAI Conference 2026 Conference Paper

MPI-Mamba: Latent Feature Fusion Mamba for Anisotropic Image Calibration and Deblurring in Magnetic Particle Imaging

Liwen Zhang
Zhaoji Miao
Yusong Shen
Zechen Wei
Hui Hui
Jie Tian

Magnetic Particle Imaging (MPI) is an innovative medical modality, providing nanomolar-scale in vivo sensitivity and radiation-free dynamic real-time detection for precision medicine. However, MPI faces a challenging problem in accurately visualizing nanoparticle distributions, where the reconstructed images with unidirectional scanning exhibit anisotropy. The anisotropy in spatial resolution leads to distortion and blurred image boundaries. Existing deep learning methods for anisotropy calibration are only limited to simulation data due to lacking of real-world MPI datasets. To address the aforementioned problems, we spent over three years designing and constructing a real-world MPI anisotropic image datasets (20,156 images) with diverse phantoms (sensitivity, resolution, vessel, shape) and animal scanning. Then, we introduce a novel Mamba-based method, MPI-Mamba, for anisotropic image calibration. Specifically, we propose a latent feature fusion state space model (LFF-SSM) block for feature fusion and leverage conditional latent diffusion model (CL-DM) branch for feature extraction. The CL-DM is performed to extract latent features in a highly compressed latent space for guiding the calibration and deblurring process. Next, we exploit the LFF-SSM to fully fuse the extracted multi-scale features to capture contextual information from the image structure, enabling the model to learn the overall distribution of signal concentration. We evaluate our method and competing methods on simulation dataset and our constructed diverse real-world MPI datasets. The results show that our proposed approach outperforms competing methods for anisotropic image calibration and deblurring.

PDF Details DOI

JBHI Journal 2026 Journal Article

TMN-LAKDE: Characterizing Sleep Instability via Prediction Intervals of Dynamic EEG Spectral Networks

Fuzhen Wei
Xiaoyu Zou
Jie Pan
Pengjun Zhao
Jiajing Song
Liwen Zhang

Sleep instability is a typical characteristic of insomnia, manifested as the inability of the brain to maintain a stable state, but its precise quantification is still challenging. We assume that sleep instability fundamentally reflects an increase in unpredictability in the evolution of brain network dynamics. To verify this, an interval prediction framework combining the temporal mobile network (TMN) and local adaptive kernel density estimation (LAKDE) is proposed to characterize the sleep instability. Specifically, TMN predicts future network states, while LAKDE module quantifies the uncertainty of these predictions by generating prediction intervals (PIs). Experiments on SIESTA and Sleep-EDF databases have shown that this method can construct well calibrated PIs. The key finding is that the PI normalized average width (PINAW) of subjects with sleep disorders is significantly higher than that of the healthy control group, validating that wider PIs are a mechanistic biomarker of sleep instability. In addition, this study further revealed a significant correlation between PINAW and traditional indicators such as number of sleep stage transitions, indicating that dynamic instability based on prediction uncertainty shares a common physiological basis with sleep fragmentation phenomena, establishing interval prediction as a paradigm for quantifying sleep instability.

Details DOI

IROS Conference 2025 Conference Paper

AUV-WTN: AUV Water Tunnel Navigation Framework with Acoustic Perturbations and Narrow Space Constraints

Haotian Zheng
Yushan Sun
Liwen Zhang
Xiaotian Wang
Jingfei Ren
Jinyu Fu

In water tunnels, autonomous navigation of autonomous underwater vehicles (AUVs) is challenging under accumulated localization errors and severe acoustic perturbations constraints. An AUV water tunnel navigation (AUV-WTN) framework is proposed to address these challenges. AUV-WTN integrates a forward-looking sonar (FLS) image segmentation method based on the refined mask R-CNN (RM R-CNN) network with real-time trajectory planning that employs the dynamic trajectory homotopy method (DTHM). RM R-CNN is optimized to combine a mixed-frequency block (MFB) along with a weighted loss function. Additionally, precise region of interest pooling (PrRoI Pooling) is combined to mitigate the impact of false targets, blurred edges, and noise on segmentation accuracy. DTHM is proposed to reduce trajectory drift by dynamically updating path generation based on segmented FLS images. Experimental results demonstrate that RM R-CNN outperforms state-of-the-art (SOTA) methods, achieving a 10. 9% improvement over Mask R-CNN in mask segmentation. The simulation platform and real AUV experiments indicate that the capability of AUV-WTN framework is effective in generating precise paths and ensuring collision-free navigation in tunnel environments.

Details

ICML Conference 2025 Conference Paper

LaRA: Benchmarking Retrieval-Augmented Generation and Long-Context LLMs - No Silver Bullet for LC or RAG Routing

Kuan Li
Liwen Zhang
Yong Jiang 0005
Pengjun Xie
Fei Huang 0002
Shuai Wang 0028
Minhao Cheng

As Large Language Model (LLM) context windows expand, the necessity of Retrieval-Augmented Generation (RAG) for integrating external knowledge is debated. Existing RAG vs. long-context (LC) LLM comparisons are often inconclusive due to benchmark limitations. We introduce LaRA, a novel benchmark with 2326 test cases across four QA tasks and three long context types, for rigorous evaluation. Our analysis of eleven LLMs reveals the optimal choice between RAG and LC depends on a complex interplay of model capabilities, context length, task type, and retrieval characteristics, offering actionable guidelines for practitioners. Our code and dataset is provided at: https: //github. com/Alibaba-NLP/LaRA

Details

NeurIPS Conference 2025 Conference Paper

WebDancer: Towards Autonomous Information Seeking Agency

Jialong Wu
Baixuan Li
Runnan Fang
Wenbiao Yin
Liwen Zhang
Zhenglin Wang
Zhengwei Tao
Ding-Chu Zhang

Addressing intricate real-world problems necessitates in-depth information seeking and multi-step reasoning. Recent progress in agentic systems, exemplified by Deep Research, underscores the potential for autonomous multi-step research. In this work, we present a cohesive paradigm for building end-to-end agentic information seeking agents from a data-centric and training-stage perspective. Our approach consists of four key stages: (1) browsing data construction, (2) trajectories sampling, (3) supervised fine-tuning for effective cold start, and (4) reinforcement learning for enhanced generalisation. We instantiate this framework in a web agent based on the ReAct format, WebDancer. Empirical evaluations on the challenging GAIA and WebWalkerQA benchmarks demonstrate the strong performance of WebDancer, achieving considerable results and highlighting the efficacy of our training paradigm. Further analysis of agent training provides valuable insights and actionable, systematic pathways for developing more capable agentic models.

PDF Details

NeurIPS Conference 2024 Conference Paper

AdaPKC: PeakConv with Adaptive Peak Receptive Field for Radar Semantic Segmentation

Teng Li
Liwen Zhang
Youcheng Zhang
Zijun Hu
Pengcheng Pi
Zongqing Lu
Qingmin Liao
Zhe Ma

Deep learning-based radar detection technology is receiving increasing attention in areas such as autonomous driving, UAV surveillance, and marine monitoring. Among recent efforts, PeakConv (PKC) provides a solution that can retain the peak response characteristics of radar signals and play the characteristics of deep convolution, thereby improving the effect of radar semantic segmentation (RSS). However, due to the use of a pre-set fixed peak receptive field sampling rule, PKC still has limitations in dealing with problems such as inconsistency of target frequency domain response broadening, non-homogeneous and time-varying characteristic of noise/clutter distribution. Therefore, this paper proposes an idea of adaptive peak receptive field, and upgrades PKC to AdaPKC based on this idea. Beyond that, a novel fine-tuning technology to further boost the performance of AdaPKC-based RSS networks is presented. Through experimental verification using various real-measured radar data (including publicly available low-cost millimeter-wave radar dataset for autonomous driving and self-collected Ku-band surveillance radar dataset), we found that the performance of AdaPKC-based models surpasses other SoTA methods in RSS tasks. The code is available at https: //github. com/lihua199710/AdaPKC.

PDF Details DOI

IJCAI Conference 2024 Conference Paper

SpecAR-Net: Spectrogram Analysis and Representation Network for Time Series

Yi Dong
Liwen Zhang
Youcheng Zhang
Shi Peng
Wen Chen
Zhe Ma

Representing temporal-structured samples is essential for effective time series analysis tasks. So far, recurrent networks, convolution networks and transformer-style models have been successively applied in temporal data representation, yielding notable results. However, most existing methods primarily focus on modeling and representing the variation patterns within time series in the time domain. As a highly abstracted information entity, 1D time series couples various patterns such as trends, seasonality, and dramatic changes (instantaneous high dynamic), it is difficult to exploit these highly coupled properties merely by analysis tools on purely time domain. To this end, we present Spectrogram Analysis and Representation Network (SpecAR-Net). SpecAR-Net aims at learning more comprehensive representations by modeling raw time series in both time and frequency domain, where an efficient joint extraction of time-frequency features is achieved through a group of learnable 2D multi-scale parallel complex convolution blocks. Experimental results show that the SpecAR-Net achieves excellent performance on 5 major downstream tasks i. e. , classification, anomaly detection, imputation, long- and short-term forecasting. Code and appendix are available at https: //github. com/Dongyi2go/SpecAR_Net.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

TARSS-Net: Temporal-Aware Radar Semantic Segmentation Network

Youcheng Zhang
Liwen Zhang
Zijun Hu
Pengcheng Pi
Teng Li
Yuanpei Chen
Shi Peng
Zhe Ma

Radar signal interpretation plays a crucial role in remote detection and ranging. With the gradual display of the advantages of neural network technology in signal processing, learning-based radar signal interpretation is becoming a research hot-spot and made great progress. And since radar semantic segmentation (RSS) can provide more fine-grained target information, it has become a more concerned direction in this field. However, the temporal information, which is an important clue for analyzing radar data, has not been exploited sufficiently in present RSS frameworks. In this work, we propose a novel temporal information learning paradigm, i. e. , data-driven temporal information aggregation with learned target-history relations. Following this idea, a flexible learning module, called Temporal Relation-Aware Module (TRAM) is carefully designed. TRAM contains two main blocks: i) an encoder for capturing the target-history temporal relations (TH-TRE) and ii) a learnable temporal relation attentive pooling (TRAP) for aggregating temporal information. Based on TRAM, an end-to-end Temporal-Aware RSS Network (TARSS-Net) is presented, which has outstanding performance on publicly available and our collected real-measured datasets. Code and supplementary materials are available at https: //github. com/zlw9161/TARSS-Net.

PDF Details DOI

ICML Conference 2023 Conference Paper

Variance Control for Distributional Reinforcement Learning

Qi Kuang
Zhoufan Zhu
Liwen Zhang
Fan Zhou

Although distributional reinforcement learning (DRL) has been widely examined in the past few years, very few studies investigate the validity of the obtained Q-function estimator in the distributional setting. To fully understand how the approximation errors of the Q-function affect the whole training process, we do some error analysis and theoretically show how to reduce both the bias and the variance of the error terms. With this new understanding, we construct a new estimator Quantiled Expansion Mean (QEM) and introduce a new DRL algorithm (QEMRL) from the statistical perspective. We extensively evaluate our QEMRL algorithm on a variety of Atari and Mujoco benchmark tasks and demonstrate that QEMRL achieves significant improvement over baseline algorithms in terms of sample efficiency and convergence performance.

Details

NeurIPS Conference 2022 Conference Paper

IM-Loss: Information Maximization Loss for Spiking Neural Networks

Yufei Guo
Yuanpei Chen
Liwen Zhang
Xiaode Liu
Yinglei Wang
Xuhui Huang
Zhe Ma

Spiking Neural Network (SNN), recognized as a type of biologically plausible architecture, has recently drawn much research attention. It transmits information by $0/1$ spikes. This bio-mimetic mechanism of SNN demonstrates extreme energy efficiency since it avoids any multiplications on neuromorphic hardware. However, the forward-passing $0/1$ spike quantization will cause information loss and accuracy degradation. To deal with this problem, the Information maximization loss (IM-Loss) that aims at maximizing the information flow in the SNN is proposed in the paper. The IM-Loss not only enhances the information expressiveness of an SNN directly but also plays a part of the role of normalization without introducing any additional operations (\textit{e. g. }, bias and scaling) in the inference phase. Additionally, we introduce a novel differentiable spike activity estimation, Evolutionary Surrogate Gradients (ESG) in SNNs. By appointing automatic evolvable surrogate gradients for spike activity function, ESG can ensure sufficient model updates at the beginning and accurate gradients at the end of the training, resulting in both easy convergence and high task performance. Experimental results on both popular non-spiking static and neuromorphic datasets show that the SNN models trained by our method outperform the current state-of-the-art algorithms.

PDF Details

IS Journal 2021 Journal Article

GC-Net: Gridding and Clustering for Traffic Object Detection With Roadside LiDAR

Liwen Zhang
Jianying Zheng
Rongchuan Sun
Yanyun Tao

The emerging intelligent transportation systems puts higher demands on the collection and analysis of the traffic data. LiDAR can provide high-precision point clouds of traffic objects, making it a promising choice for the surveillance device. This article focuses on the traffic object detection with roadside LiDAR: estimating both positions and categories of them. To overcome the challenges posed by point clouds, we propose GC-net, which is based on a three-stage pipeline, including gridding, clustering, and classification. First, we design a one-to-one mapping on raw point cloud as data preprocessing, which transforms the data structure from the graph to the grid. Then, we propose an efficient clustering algorithm: Grid- Density-Based Spatial Clustering of Applications with Noise to search the traffic objects. It exploits index information in the grid data to simplify the computational complexity. Last, we train a CNN-based classifier to categorize the found objects by extracting the local features, which performs well even the global shapes are defective. It only employs object-wise supervision, which reduces the difficulty of creating datasets. Based on the point clouds collected in real urban traffic scenarios, comparative experiences show that the proposed GC-net achieves a superior performance both in detection accuracy and computational speed, which are significant indicators for the real-time traffic surveillance systems.

Details DOI

JBHI Journal 2021 Journal Article

Multi-Focus Network to Decode Imaging Phenotype for Overall Survival Prediction of Gastric Cancer Patients

Liwen Zhang
Di Dong
Lianzhen Zhong
Cong Li
Chaoen Hu
Xin Yang
Zaiyi Liu
Rongpin Wang

Gastric cancer (GC) is the third leading cause of cancer-associated deaths globally. Accurate risk prediction of the overall survival (OS) for GC patients shows significant prognostic value, which helps identify and classify patients into different risk groups to benefit from personalized treatment. Many methods based on machine learning algorithms have been widely explored to predict the risk of OS. However, the accuracy of risk prediction has been limited and remains a challenge with existing methods. Few studies have proposed a framework and pay attention to the low-level and high-level features separately for the risk prediction of OS based on computed tomography images of GC patients. To achieve high accuracy, we propose a multi-focus fusion convolutional neural network. The network focuses on low-level and high-level features, where a subnet to focus on lower-level features and the other enhanced subnet with lateral connection to focus on higher-level semantic features. Three independent datasets of 640 GC patients are used to assess our method. Our proposed network is evaluated by metrics of the concordance index and hazard ratio. Our network outperforms state-of-the-art methods with the highest concordance index and hazard ratio in independent validation and test sets. Our results prove that our architecture can unify the separate low-level and high-level features into a single framework, and can be a powerful method for accurate risk prediction of OS.

Details DOI

IJCAI Conference 2021 Conference Paper

Non-decreasing Quantile Function Network with Efficient Exploration for Distributional Reinforcement Learning

Fan Zhou
Zhoufan Zhu
Qi Kuang
Liwen Zhang

Although distributional reinforcement learning (DRL) has been widely examined in the past few years, there are two open questions people are still trying to address. One is how to ensure the validity of the learned quantile function, the other is how to efficiently utilize the distribution information. This paper attempts to provide some new perspectives to encourage the future in-depth studies in these two fields. We first propose a non-decreasing quantile function network (NDQFN) to guarantee the monotonicity of the obtained quantile estimates and then design a general exploration framework called distributional prediction error (DPE) for DRL which utilizes the entire distribution of the quantile function. In this paper, we not only discuss the theoretical necessity of our method but also show the performance gain it achieves in practice by comparing with some competitors on Atari 2600 Games especially in some hard-explored games.

PDF Details DOI

ICML Conference 2018 Conference Paper

Tropical Geometry of Deep Neural Networks

Liwen Zhang
Gregory Naitzat
Lek-Heng Lim

We establish, for the first time, explicit connections between feedforward neural networks with ReLU activation and tropical geometry — we show that the family of such neural networks is equivalent to the family of tropical rational maps. Among other things, we deduce that feedforward ReLU neural networks with one hidden layer can be characterized by zonotopes, which serve as building blocks for deeper networks; we relate decision boundaries of such neural networks to tropical hypersurfaces, a major object of study in tropical geometry; and we prove that linear regions of such neural networks correspond to vertices of polytopes associated with tropical rational functions. An insight from our tropical formulation is that a deeper network is exponentially more expressive than a shallow network.

Details

YNICL Journal 2015 Journal Article

Brain activation during self- and other-reflection in bipolar disorder with a history of psychosis: Comparison to schizophrenia

Liwen Zhang
Esther M. Opmeer
Henricus G. Ruhé
André Aleman
Lisette van der Meer

OBJECTIVES: Reflecting on the self and on others activates specific brain areas and contributes to metacognition and social cognition. The aim of the current study is to investigate brain activation during self- and other-reflection in patients with bipolar disorder (BD). In addition, we examined whether potential abnormal brain activation in BD patients could distinguish BD from patients with schizophrenia (SZ). METHODS: During functional magnetic resonance imaging (fMRI), 17 BD patients, 17 SZ patients and 21 healthy controls (HCs) performed a self-reflection task. The task consisted of sentences divided into three conditions: self-reflection, other-reflection and semantic control. RESULTS: BD patients showed less activation in the posterior cingulate cortex (PCC) extending to the precuneus during other-reflection compared to HCs (p = 0.028 FWE corrected on cluster-level within the regions of interest). In SZ patients, the level of activation in this area was in between BD patients and HCs, with no significant differences between patients with SZ and BD. There were no group differences in brain activation during self-reflection. Moreover, there was a positive correlation between the PCC/precuneus activation during other-reflection and cognitive insight in SZ patients, but not in BD patients. CONCLUSIONS: BD patients showed less activation in the PCC/precuneus during other-reflection. This may support an account of impaired integration of emotion and memory (evaluation of past and current other-related information) in BD patients. Correlation differences of the PCC/precuneus activation with the cognitive insight in patients with BD and SZ might reflect an important difference between these disorders, which may help to further explore potentially distinguishing markers.

Details DOI