Author name cluster

Jiaxin Liu

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.

8 papers

1 author row

YNIMG Journal 2026 Journal Article

Explainable MRI radiomics of the basal ganglia and ventral midbrain distinguishes Parkinson’s disease, SWEDD, and healthy controls

Jiaxin Liu
Yuan-Zhe Li
Haomin Yang
Chong Duan
Tie-Qiang Li

Distinguishing scans without evidence of dopaminergic deficit (SWEDD) from Parkinson's disease (PD) remains challenging on routine MRI. We extracted 1284 radiomic features from T1- and T2-weighted MRI within six a priori subcortical regions of interest in the PPMI cohort. A nested cross-validation framework (inner: univariate ANOVA or Kruskal-Wallis with BH-FDR correction, mRMR, and LASSO; outer: 10-fold) was used for feature selection and classifier training. Five supervised models were evaluated, and performance was summarized by outer-fold micro- and macro-averaged AUCs with bootstrap 95% confidence intervals. Fourteen non-redundant features, primarily from the ventral midbrain, thalamus, putamen, and nucleus accumbens, were retained. The XGBoost classifier achieved a macro-AUC of 0.85 (0.76-0.91), with class-wise AUCs of 0.93 for PD, 0.79 for SWEDD, and 0.79 for healthy controls. SHAP analysis identified ventral midbrain texture heterogeneity and thalamic contrast as dominant contributors to PD prediction, while nucleus accumbens texture and putaminal shape were most informative for SWEDD. Radiomic heterogeneity on standard MRI thus captures disease-relevant patterns along a PD-SWEDD-HC continuum. Although these features are indirect surrogates of microstructure, their spatial profiles align with iron- and connectivity-related alterations reported with quantitative susceptibility and diffusion MRI. This explainable radiomics framework enables biologically coherent, multi-class discrimination between PD and SWEDD, supporting low-burden stratification and hypothesis generation for quantitative MRI studies, with planned external validation in independent cohorts to confirm generalizability.

Details DOI

AAAI Conference 2026 Conference Paper

Fair Algorithms with Probing for Multi-Agent Multi-Armed Bandits

Tianyi Xu
Jiaxin Liu
Nicholas Mattei
Zizhan Zheng

We propose a multi-agent multi-armed bandit (MA-MAB) framework to ensure fair outcomes across agents while maximizing overall system performance. For example, in a ridesharing setting where a central dispatcher assigns drivers to distinct geographic regions, utilitarian welfare (the sum of driver earnings) can be highly skewed—some drivers may receive no rides. We instead measure fairness by Nash social welfare, i.e., the product of individual rewards. A key challenge in this setting is decision-making under limited information about arm rewards (geographic regions). To address this, we introduce a novel probing mechanism that strategically gathers information about selected arms before assignment. In the offline setting, where reward distributions are known, we exploit submodularity to design a greedy probing algorithm with a constant-factor approximation guarantee. In the online setting, we develop a probing-based algorithm that achieves sublinear regret while preserving Nash social welfare. Extensive experiments on synthetic and real-world datasets demonstrate that our approach outperforms baseline methods in both fairness and efficiency.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Generalized Geometry Encoding Volume for Real-time Stereo Matching

Jiaxin Liu
Gangwei Xu
Xianqi Wang
Chengliang Zhang
Xin Yang

Real-time stereo matching methods primarily focus on enhancing in-domain performance but often overlook the critical importance of generalization in real-world applications. In contrast, recent stereo foundation models leverage monocular foundation models (MFMs) to improve generalization, but typically suffer from substantial inference latency. To address this trade-off, we propose Generalized Geometry Encoding Volume (GGEV), a novel real-time stereo matching network that achieves strong generalization. We first extract depth-aware features that encode domain-invariant structural priors as guidance for cost aggregation. Subsequently, we introduce a Depth-aware Dynamic Cost Aggregation (DDCA) module that adaptively incorporates these priors into each disparity hypothesis, effectively enhancing fragile matching relationships in unseen scenes. Both steps are lightweight and complementary, leading to the construction of a generalized geometry encoding volume with strong generalization capability. Experimental results demonstrate that our GGEV surpasses all existing real-time methods in zero-shot generalization capability, and achieves state-of-the-art performance on the KITTI 2012, KITTI 2015, and ETH3D benchmarks.

PDF Details DOI

EAAI Journal 2026 Journal Article

Hierarchical cerebral blood volume map synthesis from non-contrast magnetic resonance imaging sequences via global and local decomposition

Yihua Chen
Wangbin Ding
Guoqi Lin
Lukui Xiong
Xinhui Wang
Jiaxin Liu
Wentao Zhu
Zhaohua Lin

Cerebral blood volume (CBV) mapping is valuable for assessing brain tumor angiogenesis but relies on gadolinium-based contrast agents, which carry health risks. Generative Artificial Intelligence (AI) provides a promising non-contrast alternative; however, monolithic models have a fundamental structural limitation—they struggle to simultaneously maintain global anatomical coherence and preserve fine-grained pathological details, often leading to “texture washing” of critical vascular indicators. To overcome this, we propose HierSynth, a hierarchical framework that decomposes synthesis into separate global and local stages. A Global Synthesis Module first generates a coherent anatomical foundation. A Local Synthesis Module then refines high-frequency vascular textures exclusively in the residual domain, allowing independent optimization of fine details without compromising large-scale structure. To maintain diagnostic accuracy, we introduce a Perfusion-aware Constraint (PaC) that operates in the residual domain by supervising local refinements with perfusion-specific features (e. g. , vascular heterogeneity cues), ensuring fidelity to pathological perfusion patterns. Evaluations on internal and external datasets show HierSynth markedly outperforms state-of-the-art monolithic models, achieving higher Region of Interest (ROI) fidelity (e. g. , Structural Similarity Index (SSIM) improvement from 0. 4606 to 0. 4798). This translates clinically to better preservation of tumor vascular textures, enabling more reliable non-contrast assessment of angiogenesis. By structurally separating anatomy from pathology, HierSynth advances safer perfusion imaging and offers a generalizable paradigm for hierarchical synthesis in other diagnostic modalities requiring multi-scale detail preservation.

Details DOI

AAAI Conference 2026 Conference Paper

ReasonAct: Progressive Training for Fine-Grained Video Reasoning in Small Models

Jiaxin Liu
Zhaolu Kang

While recent multimodal models have shown progress in vision-language tasks, small-scale variants still struggle with the fine-grained temporal reasoning required for video understanding. We introduce ReasonAct, a method that enhances video reasoning in smaller models through a three-stage training process: first building a foundation with text-only reasoning, then fine-tuning on video, and finally refining with temporal-aware reinforcement learning. We build upon Temporal Group Relative Policy Optimization (T-GRPO) by incorporating temporal consistency modeling into policy optimization. We also propose a biomechanically-motivated sub-action decomposition mechanism that provides graduated rewards for constituent action phases. Through experiments on HMDB51, UCF-101, and Kinetics-400, our 3B-parameter model achieves 67.2%, 94.1%, and 78.9% accuracy respectively, demonstrating improvements of 17.9, 15.8, and 12.3 points over baselines. Ablation studies validate that our progressive training enables smaller models to achieve competitive video reasoning performance while maintaining computational efficiency.

PDF Details DOI

AAAI Conference 2025 Conference Paper

SQLFixAgent: Towards Semantic-Accurate Text-to-SQL Parsing via Consistency-Enhanced Multi-Agent Collaboration

Jipeng Cen
Jiaxin Liu
Zhixu Li
Jingjing Wang

While fine-tuned large language models (LLMs) excel in generating grammatically valid SQL in Text-to-SQL parsing, they often struggle to ensure semantic accuracy in queries, leading to user confusion and diminished system usability. To tackle this challenge, we introduce SQLFixAgent, a new consistency-enhanced multi-agent collaborative framework designed for detecting and repairing erroneous SQL. Our framework comprises a core agent, SQLRefiner, alongside two auxiliary agents: SQLReviewer and QueryCrafter. The SQLReviewer agent employs the rubber duck debugging method to identify potential semantic mismatches between SQL and user query. If the error is detected, the QueryCrafter agent generates multiple SQL as candidate repairs using a fine-tuned SQLTool. Subsequently, leveraging similar repair retrieval and failure memory reflection, the SQLRefiner agent selects the most fitting SQL statement from the candidates as the final repair. We evaluated our proposed framework on five Text-to-SQL benchmarks. The experimental results show that our method consistently enhances the performance of the baseline model, specifically achieving an execution accuracy improvement of over 3% on the Bird benchmark. Our framework also has a higher token efficiency compared to other advanced methods, making it more competitive.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

Linear Uncertainty Quantification of Graphical Model Inference

Chenghua Guo
Han Yu
Jiaxin Liu
Chao Chen
Qi Li
Sihong Xie
Xi Zhang

Uncertainty Quantification (UQ) is vital for decision makers as it offers insights into the potential reliability of data and model, enabling more informed and risk-aware decision-making. Graphical models, capable of representing data with complex dependencies, are widely used across domains. Existing sampling-based UQ methods are unbiased but cannot guarantee convergence and are time-consuming on large-scale graphs. There are fast UQ methods for graphical models with closed-form solutions and convergence guarantee but with uncertainty underestimation. We propose LinUProp, a UQ method that utilizes a novel linear propagation of uncertainty to model uncertainty among related nodes additively instead of multiplicatively, to offer linear scalability, guaranteed convergence, and closed-form solutions without underestimating uncertainty. Theoretically, we decompose the expected prediction error of the graphical model and prove that the uncertainty computed by LinUProp is the generalized variance component of the decomposition. Experimentally, we demonstrate that LinUProp is consistent with the sampling-based method but with linear scalability and fast convergence. Moreover, LinUProp outperforms competitors in uncertainty-based active learning on four real-world graph datasets, achieving higher accuracy with a lower labeling budget.

PDF Details DOI

EAAI Journal 2023 Journal Article

Research on real-time detection method of rail corrugation based on improved ShuffleNet V2

Hongjuan Yang
Jiaxin Liu
Guiming Mei
Dongsheng Yang
Xingqiao Deng
Chao Duan

Rail corrugation is a common wear mechanism of high-speed railways and subways, which can cause derailment and running noise. However, rail corrugation only has slight texture change on the rail surface, so it is difficult to detect accurately by traditional detection methods. In this paper, a real-time detection method of rail corrugation based on machine vision and a convolutional neural network is proposed, which effectively improves the accuracy and efficiency of rail corrugation detection. Combined with the gray features of each part of the image, a rail surface segmentation method based on the gray maximum value of the sliding window is also proposed. Moreover, the obtained rail surface image is clearer and the feature information of the rail surface can be completely retained, compared with the adaptive threshold segmentation and edge detection segmentation. ShuffleNet V2, a lightweight convolutional neural network, was selected as the corrugation detection model. The squeeze-and-excitation module was integrated into its basic unit to improve its channel attention, and the activation function was re-selected to make the detection have better real-time performance and accuracy. Through experimental verification, the average detection time of a single image of the improved model is 4. 01ms, and the detection accuracy is 2. 78% higher than that of the unimproved ShuffleNet V2. The research results will be beneficial to the development of the intelligent real-time detection of rail corrugation.

Details DOI