Author name cluster

Siyi Chen

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

EAAI Journal 2026 Journal Article

A new profile-agnostic sequential parametrisation method for extrusion die in transformer-based surrogate model

Jiangfeng Ding
Siyi Chen
Zhutao Shao
Zhusheng Shi
Jianguo Lin

Aluminium extrusion is widely used in manufacturing engineering components for construction, packaging, and transportation, yet its die design remains heavily dependent on time-consuming and costly trial-and-error methods informed by empirical knowledge. This research aims to expedite the most time-intensive phase, Finite Element (FE) validation phase, by proposing a sequential parametrisation of extrusion die and developing a Transformer-based surrogate model to predict the material flow speed deviation. This work consists of two distinct contributions. First, a novel sequential representation for extrusion dies is introduced, demonstrated using a T-shaped flat extrusion die. This approach eliminates the need for profile-specific parameters while preserving all geometric information of the die. The principle of the method is to decompose an extrusion die into five layers of subsections, each of which is abstracted into top/bottom views. These views are further partitioned into multiple segments arranged in a counterclockwise sequence, producing a comprehensive list of segments without any loss of information. Second, a Transformer-based autoencoder is then employed to capture the intricate interactions among these segments and generate the corresponding flow deviation plot for the current die design. Promising results on T-shaped profile have been achieved using a compact dataset with only 256 samples of FE generated data, including for similar unseen designs that have never been used for training. Comparative analyses reveal that the proposed architecture improves the accuracy of extrusion simulation predictions by 144 % (on an internal benchmark) relative to a traditional scalar-based parametrisation method using a fully connected Neural Network (NN).

Details DOI

YNIMG Journal 2026 Journal Article

Hippocampal-parietal directed connectivity mediates brain network reconfiguration between internal and external attention: An intracranial EEG study

Lizhi Yang
Huimin Huang
Xiaojun Qiao
ShengTeng Ong
Xiaoran Li
Ziyue Li
Siyi Chen
Huiqing Jia

Attention is a cornerstone of cognitive function, and understanding its neural mechanisms is of great significance for both cognitive science and clinical applications. A critical aspect of this endeavor involves elucidating how the brain's network architecture shifts between internally- and externally-directed states. However, the distinct organizational principles of neural networks in these states, as well as the pivotal brain regions and connections that mediate such transitions, remain largely unclear. To investigate these network dynamics, this study analyzed stereo-electroencephalography (SEEG) data from 17 patients with refractory epilepsy performing a modified gradual-onset continuous performance task (gradCPT) designed to induce distinct internal and external attention states. High-frequency broadband (HFB, 70-170 Hz) signals were extracted as indicators of neural activity, and neural Granger causality analysis was employed to construct effective connectivity networks between brain regions. For the effective connectivity networks, we systematically applied modular analysis to quantify network segregation, node role classification to identify hub regions, and machine learning methods to evaluate the discriminative power of the identified connectivity differences. The results showed that the external attention state exhibited significantly stronger global causal connectivity and a topological profile dominated by connector hubs. In contrast, the internal attention state displayed higher modularity and a prevalence of peripheral nodes, reflecting a segregated network architecture. Eight pairs of brain region connections showed significant differences between the two states, primarily involving the parietal-temporal network. A support vector machine (SVM) classifier achieved 77.8% accuracy in distinguishing attention states under cross-subject conditions using the identified directed connectivity features, demonstrating the discriminative power of network differences. Feature importance analysis identified the intrinsic dynamics of the hippocampus (HIP) and its directed outflow to the middle temporal gyrus (MTG) as the most significant discriminative features. Consequently, the hippocampus operates in concert with temporal and parietal regions to mediate these transitions, suggesting that flexible cognitive control depends on the dynamic coupling between memory systems and cortical networks. These results provide potential neural biomarkers for attention-related disorders and advance our mechanistic understanding of how the brain adaptively organizes information flow to meet varying cognitive demands.

Details DOI

YNIMG Journal 2025 Journal Article

Distinct hippocampus codes for contextual cueing: learning contexts and their predictive associations with targets in visual search

Siyi Chen
Si Cheng
Thomas Geyer
Hermann J. Müller
Zhuanghua Shi

Humans can learn to exploit repeated distractor arrangements to optimize attentional selection of targets, producing contextual facilitation. The hippocampus is thought to support context representations acquired from repeatedly searching a given scene layout. However, it remains unclear whether the hippocampus primarily encodes context-target associations, in which the distractor layout directly predicts the target location, or whether it additionally encodes associations among distractors, enabling target prioritization indirectly via context suppression. To examine the neural mechanisms of contextual learning, we combined functional magnetic resonance imaging with a two-phase visual search paradigm: Phase 1 presented predictive (repeated) distractor layouts with consistent target locations, affording contextual cueing; Phase 2 rendered these layouts non-predictive by randomizing target locations, fostering context suppression. Contextual facilitation was compared against a baseline of non-repeated arrangements. We found that both context-target (Phase 1) and distractor-distractor (Phase 2) associations were reliably decoded from the hippocampus using correlation-based multi-voxel pattern analysis. A functional dissociation emerged along the hippocampal axis: anterior and posterior hippocampal regions identified in the whole-brain univariate analyses exhibited relatively greater contributions to Phase 1 context-target and Phase 2 distractor-distractor associations, respectively, indicating their stronger involvement in the corresponding memory representations. Connectivity modeling showed the temporoparietal junction (TPJ) interacted with the hippocampus in different ways depending on context predictivity. These findings indicate anatomically separable hippocampal circuits represent predictive context-target and non-predictive distractor-distractor relations, with their attentional effects gated by the TPJ. SIGNIFICANCE STATEMENT: Although the hippocampus supports the encoding and retrieval of recurrent spatial distractor-target relations in visual search, its distinct roles in representing context-target relations (associating the target location with the configuration of distractors) versus sole-context (distractor-distractor) configurations has not been dissociated. The present study decoded both forms of contextual representation in the hippocampus: the anterior hippocampus preferentially encoded context-target associations, whereas the posterior hippocampus maintained sole-context memory. The signals generated by these distinct hippocampal regions regulate how the target is prioritized for attentional selection, with the temporoparietal junction (TPJ) dynamically adjusting the mode of prioritization in response to the learnt configural structure and how reliably it predicts the target location.

Details DOI

NeurIPS Conference 2025 Conference Paper

FlowDAS: A Stochastic Interpolant-based Framework for Data Assimilation

Siyi Chen
Yixuan Jia
Qing Qu
He Sun
Jeffrey Fessler

Data assimilation (DA) integrates observations with a dynamical model to estimate states of PDE-governed systems. Model-driven methods (e. g. , Kalman Filter, Particle Filter) presuppose full knowledge of the true dynamics, which is not always satisfied in practice, while purely data-driven solvers learn a deterministic mapping between observations and states and therefore miss the intrinsic stochasticity of real processes. Recently, score-based diffusion models have shown promise for DA by learning a global diffusion prior to represent stochastic dynamics. However, their one-shot generation lacks stepwise physical consistency and struggles with complex stochastic processes. To address these issues, we propose FlowDAS, a generative DA framework that employs stochastic interpolants to learn state transition dynamics through step-by-step stochastic updates. By incorporating observations into each transition, FlowDAS can produce stable, measurement-consistent forecasts. Experiments on Lorenz-63, Navier–Stokes super-resolution/sparse-observation scenarios, and large-scale weather forecasting—where dynamics are partly or wholly unknown—show that FlowDAS surpasses model-driven methods, neural operators, and score-based baselines in accuracy and physical plausibility. Our implementation is available at https: //github. com/umjiayx/FlowDAS.

PDF Details

JBHI Journal 2025 Journal Article

Localized Intra- and Inter-Tumoral Heterogeneity for Predicting Treatment Response to Neoadjuvant Chemotherapy in Breast Cancer

Yinhao Liang
Wenjie Tang
Qingcong Kong
Ting Wang
Jianjun Zhang
Wing W. Y. Ng
Siyi Chen
Ying Li

This study proposes a novel method for extracting breast cancer tumor heterogeneity descriptors to non-invasively predict whether pathological complete response (pCR) can be achieved after neoadjuvant chemotherapy (NAC). These localized descriptors extract corresponding heterogeneity features for different radiomic features and are able to capture tumor characteristics at various localization levels. These descriptors also capture tumor heterogeneity both at the individual tumor level and across the whole dataset, providing decision-making models with features that are both more effective and interpretable. We validated the effectiveness of the proposed features with the Kolmogorov-Arnold network (KAN) across multiple centers, yielding an AUC of 0. 92 when combined with pathological features and demonstrating good performance in external datasets (AUCs of 0. 84 and 0. 81). Additionally, we transform the best model into a symbolic formula to intuitively explain the machine learning model's prediction process, showing how factors such as age, HER2, Ki-67 and heterogeneity influence the prediction. The symbolized model is consistent with the experience of clinical experts, which enhances users' confidence in deep models. The experimental results show that our proposed features and method outperform classical heterogeneity features and end-to-end neural networks with a small additional computational cost.

Details DOI

NeurIPS Conference 2025 Conference Paper

Understanding Representation Dynamics of Diffusion Models via Low-Dimensional Modeling

Xiao Li
Zekai Zhang
Xiang Li
Siyi Chen
Zhihui Zhu
Peng Wang
Qing Qu

Diffusion models, though originally designed for generative tasks, have demonstrated impressive self-supervised representation learning capabilities. A particularly intriguing phenomenon in these models is the emergence of unimodal representation dynamics, where the quality of learned features peaks at an intermediate noise level. In this work, we conduct a comprehensive theoretical and empirical investigation of this phenomenon. Leveraging the inherent low-dimensionality structure of image data, we theoretically demonstrate that the unimodal dynamic emerges when the diffusion model successfully captures the underlying data distribution. The unimodality arises from an interplay between denoising strength and class confidence across noise scales. Empirically, we further show that, in classification tasks, the presence of unimodal dynamics reliably reflects the diffusion model’s generalization: it emerges when the model generate novel images and gradually transitions to a monotonically decreasing curve as the model begins to memorize the training data.

PDF Details

JBHI Journal 2025 Journal Article

XRadNet: A Radiomics-Guided Breast Cancer Molecular Subtype Prediction Network With a Radiomics Explanation

Yinhao Liang
Wenjie Tang
Jianjun Zhang
Ting Wang
Wing W. Y. Ng
Siyi Chen
Kuiming Jiang
Xinhua Wei

In this work, we propose a radiomics-guided neural network, XRadNet, for breast cancer molecular subtype prediction. XRadNet is a two-head neural network, with one for predicting molecular subtypes and the other for approximating radiomic features. In addition, a training scheme with radiomics guidance is proposed to improve performance. First, we conduct a series of experiments to test the radiomic feature learning capacity of different neural networks, which determines the backbone of XRadNet. Moreover, significant radiomic features are also determined according to radiomics and prior knowledge. XRadNet is subsequently pretrained in a self-supervised manner. The pretraining uses synthetic samples to train the backbone and radiomic feature regression head. This mitigates the impact of an insufficient number of samples. Finally, XRadNet is fine-tuned with a downstream real-world dataset by enabling all heads. Furthermore, a logistic regression is built with radiomic features and learned features, which provides a new way to interpreting the trained model with concepts familiar to radiologists. The experimental results show that XRadNet effectively predicts the four molecular subtypes of breast cancer. These results also demonstrate that the proposed training scheme yields better or competitive performance than those models pretrained on ImageNet or medical datasets.

Details DOI

YNIMG Journal 2024 Journal Article

Attentional control influence habituation through modulation of connectivity patterns within the prefrontal cortex: Insights from stereo-EEG

Huimin Huang
Rui Li
Xiaojun Qiao
Xiaoran Li
Ziyue Li
Siyi Chen
Yi Yao
Fengpeng Wang

Attentional control, guided by top-down processes, enables selective focus on pertinent information, while habituation, influenced by bottom-up factors and prior experiences, shapes cognitive responses by emphasizing stimulus relevance. These two fundamental processes collaborate to regulate cognitive behavior, with the prefrontal cortex and its subregions playing a pivotal role. Nevertheless, the intricate neural mechanisms underlying the interaction between attentional control and habituation are still a subject of ongoing exploration. To our knowledge, there is a dearth of comprehensive studies on the functional connectivity between subsystems within the prefrontal cortex during attentional control processes in both primates and humans. Utilizing stereo-electroencephalogram (SEEG) recordings during the Stroop task, we observed top-down dominance effects and corresponding connectivity patterns among the orbitofrontal cortex (OFC), the middle frontal gyrus (MFG), and the inferior frontal gyrus (IFG) during heightened attentional control. These findings highlighting the involvement of OFC in habituation through top-down attention. Our study unveils unique connectivity profiles, shedding light on the neural interplay between top-down and bottom-up attentional control processes, shaping goal-directed attention.

Details DOI

NeurIPS Conference 2024 Conference Paper

Exploring Low-Dimensional Subspace in Diffusion Models for Controllable Image Editing

Siyi Chen
Huijie Zhang
Minzhe Guo
Yifu Lu
Peng Wang
Qing Qu

Recently, diffusion models have emerged as a powerful class of generative models. Despite their success, there is still limited understanding of their semantic spaces. This makes it challenging to achieve precise and disentangled image generation without additional training, especially in an unsupervised way. In this work, we improve the understanding of their semantic spaces from intriguing observations: among a certain range of noise levels, (1) the learned posterior mean predictor (PMP) in the diffusion model is locally linear, and (2) the singular vectors of its Jacobian lie in low-dimensional semantic subspaces. We provide a solid theoretical basis to justify the linearity and low-rankness in the PMP. These insights allow us to propose an unsupervised, single-step, training-free LO w-rank CO ntrollable image editing (LOCO Edit) method for precise local editing in diffusion models. LOCO Edit identified editing directions with nice properties: homogeneity, transferability, composability, and linearity. These properties of LOCO Edit benefit greatly from the low-dimensional semantic subspace. Our method can further be extended to unsupervised or text-supervised editing in various text-to-image diffusion models (T-LOCO Edit). Finally, extensive empirical experiments demonstrate the effectiveness and efficiency of LOCO Edit. The code and the arXiv version can be found on the project website.

PDF Details DOI

YNIMG Journal 2020 Journal Article

Tracking the completion of parts into whole objects: Retinotopic activation in response to illusory figures in the lateral occipital complex

Siyi Chen
Ralph Weidner
Hang Zeng
Gereon R. Fink
Hermann J. Müller
Markus Conci

Illusory figures demonstrate the visual system’s ability to integrate separate parts into coherent, whole objects. The present study was performed to track the neuronal object construction process in human observers, by incrementally manipulating the grouping strength within a given configuration until the emergence of a whole-object representation. Two tasks were employed: First, in the spatial localization task, object completion could facilitate performance and was task-relevant, whereas it was irrelevant in the second, luminance discrimination task. Concurrent functional magnetic resonance imaging (fMRI) used spatial localizers to locate brain regions representing task-critical illusory-figure parts to investigate whether the step-wise object construction process would modulate neural activity in these localized brain regions. The results revealed that both V1 and the lateral occipital complex (LOC, with sub-regions LO1 and LO2) were involved in Kanizsa figure processing. However, completion-specific activations were found predominantly in LOC, where neural activity exhibited a modulation in accord with the configuration’s grouping strength, whether or not the configuration was relevant to performing the task at hand. Moreover, right LOC activations were confined to LO2 and responded primarily to surface and shape completions, whereas left LOC exhibited activations in both LO1 and LO2 and was related to encoding shape structures with more detail. Together, these results demonstrate that various grouping properties within a visual scene are integrated automatically in LOC, with sub-regions located in different hemispheres specializing in the component sub-processes that render completed objects.

Details DOI