Research paper pages https://codezx6.github.io/ 2026-09-27T00:00:00Z Research paper pages URMDet-SimFire: Trustworthy multimodal detection with uncertainty and reliability modeling for fire and smoke analyticshttps://codezx6.github.io/papers/urmdet-simfire.html2026-08-07T00:00:00Z2026-09-27T00:00:00ZZhonghua DangXu ZhangURMDet-SimFire is an uncertainty- and reliability-aware multimodal object detection framework for fire and smoke that reweights each input modality by its estimated reliability and calibrates each detection's confidence by its uncertainty. In simulation only, it raised F1 from 0.4348 (vision-only baseline) to 0.4638. DUET: Unified Dual-Space Emotion Control for Diffusion and Flow-Matching Driven Text-to-Speechhttps://codezx6.github.io/papers/duet.html2026-05-20T00:00:00Z2026-09-27T00:00:00ZXu ZhangLongbing CaoZhangkai WuDUET adds emotion control to frozen diffusion and flow-matching TTS models by steering hidden states along a linearly decodable emotion direction and refining the mel estimate with emotion-recognizer gradients passed through a differentiable vocoder. With GradTTS on ESD it reaches 75.5% average emotion accuracy (strongest supervised baseline: 46.8%). PhysioSER: Learning Physiology-Informed Vocal Spectrotemporal Representations for Speech Emotion Recognitionhttps://codezx6.github.io/papers/physioser.html2026-02-03T00:00:00Z2026-09-27T00:00:00ZXu ZhangLongbing CaoRunze YangZhangkai WuPhysioSER complements a frozen self-supervised (SSL) backbone with a compact physiology-informed branch that encodes vocal amplitude and phase features with quaternion convolutions, for speech emotion recognition. With frozen WavLM on CREMA-D, it raises weighted accuracy from 69.69% to 75.20%. DSTCN: Exploiting dynamic spatio-temporal correlations for origin-destination demand predictionhttps://codezx6.github.io/papers/dstcn.html2025-10-24T00:00:00Z2026-09-27T00:00:00ZYongshun GongPiao YuXu ZhangXinxin ZhangXiushan NieHaoliang SunDSTCN (Dynamic Spatio-Temporal Correlation Network) forecasts origin-destination demand matrices with three modules: Glstm2D for origin- and destination-side demand trends, Simformer for Transformer-based spatial similarity across the OD matrix, and FF-TM for feature fusion and temporal modeling. On NYC-TOD2018 its MAE of 1.469 is 3.04% below the best baseline. S2CMEN: A Mutually Enhancement Network for Superpixel Segmentation and Classification of Hyperspectral Imagehttps://codezx6.github.io/papers/mutual-enhancement-hsi.html2025-09-11T00:00:00Z2026-09-27T00:00:00ZMengxin CaoYongmin LiXu ZhangGuixin ZhaoGuohua LvAimei DongJinyong ChengWei LiXiangjun DongS²CMEN (S2CMEN) is a hyperspectral image classification network in which superpixel segmentation and classification enhance each other through a unified loss, combining superpixel-based global spatial context with Spectral-Swin Transformer spectral features. It reaches 97.38% overall accuracy on Indian Pines. BiST-IF: Enhancing origin–destination flow prediction via bi-directional spatio-temporal inference and interconnected feature evolutionhttps://codezx6.github.io/papers/bist-if.html2024-11-22T00:00:00Z2026-09-27T00:00:00ZPiao YuXu ZhangYongshun GongJian ZhangHaoliang SunJunjie ZhangXinxin ZhangYilong YinBiST-IF predicts origin–destination (OD) flows between metro stations or urban areas by correcting delayed recent OD matrices, applying bi-directional origin/destination attention, and fusing arrival-side (Out-OD) flows through an attention-based mutual information mechanism. It lowers MAE by an average of 7.55% on HZMetro relative to the best baseline. MR-UFP: Enhancing urban flow prediction via mutual reinforcement with multi-scale regional informationhttps://codezx6.github.io/papers/mr-ufp.html2024-11-16T00:00:00Z2026-09-27T00:00:00ZXu ZhangMengxin CaoYongshun GongXiaoming WuXiangjun DongYing GuoLong ZhaoChengqi ZhangMR-UFP predicts the inflow and outflow of city grid regions, even with limited training data, by pre-training encoders with spatial-temporal masking and contrastive learning and training flow prediction jointly with a multi-scale region-classification task. On full TaxiBJ and BikeNYC it beats all 13 baselines (RMSE 14.32 and 4.45). Automatic visual recognition for leaf disease based on enhanced attention mechanismhttps://codezx6.github.io/papers/leaf-disease-attention.html2024-11-04T00:00:00Z2026-09-27T00:00:00ZYumeng YaoXiaodun DengXu ZhangJunming LiWenxuan SunGechao ZhangA tomato leaf disease detector that adds the DyHead attention module to YOLOv4-tiny and trains it with a Focaler-SIoU box loss. On PlantDoc tomato leaf images it reaches 93.64% mAP, 10.3 percentage points above the YOLOv4-tiny baseline. S3CFSL: Spatial-Spectral–Semantic Cross-Domain Few-Shot Learning for Hyperspectral Image Classificationhttps://codezx6.github.io/papers/spatial-spectral-semantic-fsl.html2024-07-29T00:00:00Z2026-09-27T00:00:00ZMengxin CaoXu ZhangJinyong ChengGuixin ZhaoWei LiXiangjun DongS3CFSL classifies new hyperspectral scenes from five labelled samples per class by transferring knowledge from the labelled Chikusei scene. It combines a cross-spatial–spectral transformer, Gaussian feature denoising and semantic-enhanced domain alignment, and reports 98.52% overall accuracy on Pavia Centre, 88.79% on Salinas and 77.63% on Houston. ST-FCL: Spatio-temporal fusion and contrastive learning for urban flow predictionhttps://codezx6.github.io/papers/st-csl.html2023-10-21T00:00:00Z2026-09-27T00:00:00ZXu ZhangYongshun GongChengqi ZhangXiaoming WuYing GuoWenpeng LuLong ZhaoXiangjun DongST-FCL predicts grid-level urban inflow and outflow by fusing temporal and spatial views, learned through contrastive pretraining, with an external-factor view. On the full TaxiBJ dataset it reaches RMSE 14.71, against 15.41 for the best baseline, ATFM. MC-STL: Mask- and Contrast-Enhanced Spatio-Temporal Learning for Urban Flow Predictionhttps://codezx6.github.io/papers/mcstl.html2023-10-21T00:00:00Z2026-09-27T00:00:00ZXu ZhangYongshun GongXinxin ZhangXiaoming WuChengqi ZhangXiangjun DongMC-STL pre-trains two encoders for urban flow prediction: a ViT encoder learns to reconstruct regions masked at different timestamps, and its attention weights also build a GCN adjacency matrix; a global-local cross-attention encoder learns a temporal-order contrastive task. It reaches RMSE 14.53 on full TaxiBJ.