Quick Navigation

1. Agents, Robotics and Multi-Agent Systems (Sessions F1.1–F1.6, 43 papers)

Session F1.1 (8 papers)Nov 19 (Thu), 8:30 AM – 10:30 AMRoom R205

1. ManeuverPark : Learning Maneuver Sets for End-to-End Autonomous Parking in Dynamic Scenes Hualong Zhang, Ammar Hawbani, Ping Liu, Liang Zhao and Jiantao Wang [ID 153]
2. An Empirical Study of Adaptive Self-Confidence Scheduling in Mixed Opinion Dynamics: Rule-based vs. Learning-based Policies Tianze Gao, Jinchuan Zhang, Chong Mu, Lizong Zhang, Sicong Liu and Lei Gao [ID 283]
3. Co-Evolutionary Optimization of Warehouse Layouts and Multi-Agent Path Finding Policies Xiangjie Luo, Miyuki Koshimura and Makoto Yokoo [ID 590]
4. DynaWM: A Base-VLA-Guided World Foundation Model for Moving-Object Manipulation Chongkei Chang and Zhidong Deng [ID 805]
5. Hydra: Dynamic Hydrological Reasoning Agents for Disaster Forecasting Duy-Thanh-Long Van, Vinh-Tiep Nguyen, Tu-Anh Nguyen-Hoang and Ngan Luu-Thuy Nguyen [ID 862]
6. MAGRO: A Multi-Agent Graph Reasoning GRPO Optimization Framework for Complex Reasoning in Large Language Models Benhui Cao and Jing Xiao [ID 933]
7. Diffusion Policy-Based Autonomous Material Pile Stacking for Unmanned Loaders Guangda Chen, Shunyi Yao, Jiyuan Ru, Lin Cheng, Zhiwen Zhang, Yingfeng Chen, Keke Tang and Jianmin Ji [ID 1232]
8. DistCoT-Agent: Empowering Autonomous Software Engineering via Trajectory-Level Preference Optimization Guanyu Lin, Jing Liu, Xuyang Song and Li Han [ID 1313]

Session F1.2 (8 papers)Nov 19 (Thu), 8:30 AM – 10:30 AMRoom R208

1. A Safe Weather Adaptive DRL Approach for Unsignalized Intersections Shiyu Liu, Ammar Hawbani, Redhwan Algabri and Liang Zhao [ID 154]
2. Epistemic-Stance Poisoning: A Stealthy Residual Channel that Derails Multi-Hop LLM Agents Xingfu Zhou and Pengfei Wang [ID 487]
3. ConflictBench-Agent: A Diagnostic Benchmark and Trust-Conflict Formalism for Evidence Arbitration in LLM Agents Zhurun Zhang [ID 1401]
4. Communication-Constrained Multi-UGV Rescue Task Allocation with UAV-Assisted Broadcasting Runa Song, Xianchang Wang and Jianjie Yang [ID 1571]
5. Glance-then-Task: A Boundary Study of Test-Time Anchoring for Multi-Constraint LLM Tasks Guoqing Ma, Jia Zhu, Jiawei Shen, Yidan Liang and Jingjiang Liu [ID 1701]
6. Sustaining Multiple In-Flight Tasks: A Single Agent for Interleaved Execution Xubo Cai, Duo Li, Lu Chen, Peng Yi and Cheng Yang [ID 1714]
7. SafeAgentBench: A Benchmark for Safe Task Planning of Embodied VLM Agent Sheng Yin, Xianghe Pang, Yuanzhuo Ding, Menglan Chen, Yutong Bi, Yichen Xiong, Wenhao Huang, Zhen Xiang, Jing Shao and Siheng Chen [ID 1815]
8. Adaptive Skill Granularity for Coding Agents: Runtime Routing and Macro-to-Micro Recovery Dingxuan Zhang, Yanhui Jiang, Jun Wang and Hefeng Zhou [ID 1817]

Session F1.3 (8 papers)Nov 19 (Thu), 8:30 AM – 10:30 AMRoom R210

1. PMO-SAC: A Hybrid Reinforcement Learning Framework for Human-Centric Vehicle Dispatching with Bilateral Preferences Jialiang Bao, Hui Zeng, Jia Yu, Xuebo Cheng and Ni Liu [ID 295]
2. An Efficient Customized Policy Transfer Framework for Heterogeneous Multi-Agent Reinforcement Learning Kun Yang, Zhemin Li, Ziyi Su, Zheng Xie and Yiping Song [ID 420]
3. SA-VDN: Slack-Aware Value Decomposition for Heterogeneous Multi-GPU Medical Imaging Scheduling Xinmei Wang, Jing Ge, Jing Chen, Yinglong Wang, Ying Guo and Xiaoming Wu [ID 432]
4. Trans-DDQN: Trading Volatility through Straddle Options with Dual-Attention Q-Learning Keyi Feng, Yujing Liu and Zhiguo Tao [ID 756]
5. From Evaluation to Improvement: Multi-Agent Pedagogical Feedback of LLM Tutors Towards Improving Student Learning in Dialogues Sebastián A. Burgos-Martínez, Wenbin Gan and Yuan Sun [ID 813]
6. Inference-based Imperfect Information Monte-Carlo Simulations for Indian Rummy Santhosh Kumar Guguloth and Shivaram Kalyanakrishnan [ID 1007]
7. A Continual Reinforcement Learning Energy Management Agent for Fuel Cell Hybrid Electric Vehicles Zhiwu Huang, Zhaosheng Qiu, Yue Wu, Jun Peng and Xiaoyong Zhang [ID 1345]
8. Position: We Should Evaluate Agentic Memory as Inference, not as a Tape Recorder Yiannis Hadjiyianni, Panagiotis Michelakis, Emmanouil Neofotistos, Jing Jiang, Dimitrios Stamoulis and Jie Liu [ID 1716]

Session F1.4 (7 papers)Nov 19 (Thu), 8:30 AM – 10:30 AMRoom R212

1. Think Fast, Think Slow in Social Simulation: A Dual-Process Cognitive Routing Framework for LLM Agents Yuhan Zhang, Haojie Wu, Haowei Li, Jie Bao and Yong Liao [ID 356]
2. TeamCoder: Hierarchical Multi-Agent Code Generation via Cognitive Funneling Wentao Xu, Jianfeng Huang, Chen Liang and Qing Su [ID 889]
3. ClusterMem: A Hierarchical Memory Architecture for Long-Term Conversational Agent Haibo Chao, Zhen Zeng, Zhendao Chen, Yan Pan and Ning Tan [ID 908]
4. CMASE: Counterfactual Multi-Head Attention for Side-Effect-Aware Multi-Agent Coordination Xiangling Yang, Shuli Ai and Wenjun Ma [ID 1027]
5. Mindfulness Attitudes as Intermediate Representations for Safety-Aware Multi-Agent LLM Psychological Support Shihong Chen, Bin He and Ru Yan [ID 1181]
6. HiMAP: Hierarchical Intent-Grounded Multi-Agent Planning for Executable Simulation Scenario Generation Xing Niu and Minchi Kuang [ID 1576]
7. GoTopoGuard: Topology-Aware Risk Monitoring for Rare Structural Failures in Neural Go Engines Yixi Chen [ID 694]

Session F1.5 (6 papers)Nov 19 (Thu), 10:50 AM – 12:20 PMRoom R205

1. Structured State Management with Feedback Augmentation for Web Agents Jingyao Xu, Duanyang Yuan, Xirui Liu, Maolin He and Sihang Zhou [ID 696]
2. VOTQA: Verifiable Operational Tables for Question Answering over Complex Spreadsheets Jingqi Zuo, Feng Chen, Wenjia Wu, Hong Li, Changjun Wang and Chie Sakuragi [ID 715]
3. CARTA: Coordinate-Anchored Graph Reasoning and Topology-Aware Agent for Semi-Structured Table Question Answering Yunhao Lu, Xintao Jiao and Haolan Yang [ID 872]
4. Moral Utility for Stable Coalitions in Cooperative Games Takayuki Ito [ID 1508]
5. GAFT: Geo-Anchored Fine-Tuning for Hazard Identification from Rare Failures Yanran Xu, Chuanhang Qiu, Yue Wang, Wenbo Wu and Zhaoxing Li [ID 1653]
6. CLEAR-Mem: Claim-Level Evidence and Authority Regulation for Rolling Long-Context Compaction Jianfeng Lan [ID 952]

Session F1.6 (6 papers)Nov 19 (Thu), 10:50 AM – 12:20 PMRoom R208

1. FailSafe-Agent: A Controlled Study of State-Verified Tool Commitment Yanxiao Li, Junhao Wei, Yifu Zhao, Haochen Li, Xu Yang and Yapeng Wang [ID 984]
2. TRACE-LLM: Auditable Constraint Control and Guarded Revision for Evidence-Grounded EAP Response Generation Chenxi Liu and Raymond Lee [ID 1061]
3. Effect Is Not Control: Evaluating Precise Tool-Call Reachability in Tool-Augmented LLM Agents Jiaming Liu and Hao Wang [ID 1267]
4. Echo: Contradiction-Preserving Self-Model Revision for LLM Agents Zicai Cui [ID 1511]
5. DECC-RAG: Debate-Guided Evidence Chain Completion for Retrieval-Augmented Generation Jun Zheng, Yubin Bao, Fangling Leng, Zhaohui Fang, Tiancheng Zhang, Kexin Wu and Ge Yu [ID 1529]
6. SpiralSQL: Ambiguity-Driven Database Probing and Progressive Generation for Text-to-SQL Qingshuo Lyu, Jiangyi Yin, Zhao Li, Zhuojun Jiang, Xudong Yao and Qingyun Liu [ID 1620]

2. AI in Education (Sessions F2.1, 5 papers)

Session F2.1 (5 papers)Nov 20 (Fri), 3:20 – 4:35 PMRoom R316

1. Subjective-Graph LLM Agents for Simulating Uncertainty in Classroom Social Perception Jinming Yang, Xinyu Jiang, Xinshan Jiao and Xinping Zhang [ID 1114]
2. Within-Student Evolution of LLM Help-Seeking in Programming: Clusters, Trends, and Task Effects Zhenhai He and Shaoming Chai [ID 728]
3. Learning When to Intervene: A Theory-Grounded Trigger Model for AI Facilitators in Knowledge Building Communities Zhenhai He, Shaoming Chai and Tongwu Jiang [ID 734]
4. Beyond Random Noise: MorphoGEC for Hindi Grammatical Error Correction Sourava Kumar Behera and Rohit Saluja [ID 1296]
5. TD-SCR: Time-Difficulty Soft Candidate Regulation for Offline Learning Path Recommendation Hongtao Yu, Yiping Wen, Zhiyuan Ai, Xin Yan and Jiayi Ma [ID 1379]

3. Big Data Analytics and Internet of Things (Sessions F3.1–F3.4, 23 papers)

Session F3.1 (8 papers)Nov 20 (Fri), 8:30 AM – 10:30 AMRoom R321

1. WG-TAP-LLM: A Wavelet-Graph Traffic-Aware Prompt Large Language Model for Traffic Flow Forecasting Junhong Zheng, Huaixiang Zhang and Yu Chen [ID 165]
2. LSG-PIR: Detection-Decoupled Privacy Image Recognition for IoT Devices via Local Semantic Guidance Shaopeng Zhou, Feng Zhang, Xing Wang, Song Liu, Si Chen, Kezhang Lin, Longlong Zhu, Chunming Wu and Bin Wang [ID 565]
3. LSRR: A Lightweight Semantic Residual Refinement Framework for Non-Stationary Cellular Traffic Forecasting Qingzhong Li, Yue Hu, Qingchang Ma, Zhou Long, Zongyu Zhang, Tingting Guan, Xinjun Pei, Hui Ma and Yajun Zhang [ID 1161]
4. Attentive Light: Lightweight IoT Edge Attack Detection via Sparse Spatio-Temporal Sequence Modeling Lihang Dai, Lisheng Huang and Yunhai Lan [ID 1211]
5. UPVNet: Uncertainty-Aware Probabilistic Photovoltaic Power Forecasting Network Shouyuan Tang, Lingcui Sun, Jinbao Xiang, Mengjun Wang, Yichen Liu, Haoran Xu, Hao Chen and Shiyu Zhu [ID 1272]
6. FDIMixer: Frequency-Decoupled Inverted Channel Attention for Efficient High-Dimensional Time Series Forecasting Qun Chen and Yumin Dong [ID 1393]
7. ChronoMem-Former: Memory-Routed Short-Context Transformers for Long-Term Multivariate Time Series Forecasting Yujie Yao and Jun Long [ID 1775]
8. CSST: Component-Stability Selective Training for Long-Term Time-Series Forecasting Hao Yang [ID 1859]

Session F3.2 (5 papers)Nov 20 (Fri), 3:20 – 4:35 PMRoom R205

1. Tri-Perspective Hypergraph Programming Knowledge Tracing Jianguo Pan, Yuchen Wan, Zhengyang Dong, Zihao Lian, Zijun Wang, Peichi Zhou and Yuan Yang [ID 310]
2. PPAL: Prototypical Precursor-Aware Learning for Unsupervised Time-Series Anomaly Prediction Pochen Lai and Vincent S. Tseng [ID 673]
3. Sparse Soft Subspace Spectral Clustering (4SC) Abdoul Wahab Diallo, Mory Ouattara, Ndèye Niang and Mamadou Bousso [ID 979]
4. A Heterogeneous Hyperedge Prediction Approach to Model Orchestration in Edge Intelligence Yiting You, Yan Gao, Jianfa Wang and Ruizhong Du [ID 1312]
5. CipherNews: Efficient and Private Inference for Content-Based Neural News Recommendation Parhat Abla, Junguang Yao, Lihua Cai, Zhihong Deng and Taotao Li [ID 1325]

Session F3.3 (5 papers)Nov 20 (Fri), 3:20 – 4:35 PMRoom R208

1. High-Efficiency Split Computing for Cooperative Edge Systems: A Variational Feature Measurement Bottleneck Hailin Zhong and Shengxin Zhu [ID 1559]
2. GMSPRec: Graph-Frequency Spectral Preference Learning for Session-Based Recommendation Chuhui Wu and Fengfan Wu [ID 1671]
3. Road-Network Origin-Destination Travel Time Estimation for Last-mile Delivery via Spatial Grid Sampling and Cascaded Learning Pan Yuhong, Xing Ershang and Han Kunpeng [ID 589]
4. PS-BiGAN: Point-Sequence Fusion for One-Class Misbehavior Detection in Vehicular Networks Qifeng Xiao, Jianming Cui and Ming Liu [ID 688]
5. From Mixed Readings to Reliable NO2 Estimates: Mechanism-Informed Historical Contribution Modeling for Mobile Sensor Calibration Yuzhuang Pian, Taiyu Wang, Qihang Li, Ruhui Cao and Yonghong Liu [ID 709]

Session F3.4 (5 papers)Nov 20 (Fri), 3:20 – 4:35 PMRoom R210

1. Device-wise Conformal False-Alarm Calibration for Federated IoT Intrusion Detection Renchao Li and Aijiao Cui [ID 1168]
2. Curriculum-Guided Graph Reinforcement Learning for Vehicular Task-Chain Scheduling Yonghui Ye, Yan Gao, Shaoyuan Huang, Yan Liu and Ruizhong Du [ID 1269]
3. HAPGUARD: Joint Attacker Identification and Resilient Aggregation for Federated IoT Intrusion Detection Muhammad Samiullah, Muhammad Owais, Kainat Ghazanfar and Ahmad Chaddad [ID 1598]
4. TAR-FL: Auditable Trigger-Aware Reputation Calibration for Non-IID Federated Learning Ziliang Zhang, Jianping Yu, Xiaojun Cao and Jiajun Huang [ID 1750]
5. Cascaded Kolmogorov-Arnold Networks for Interpretable Aircraft Trajectory Prediction under Complex Wind Fields Yifu Zhang, Lang Huang, Jinzhu Hu, Youcao Wu, Yunli Peng, Chun Shen and Jianbing Li [ID 1851]

4. Bioinformatics and Biomedical AI (Sessions F4.1–F4.3, 22 papers)

Session F4.1 (8 papers)Nov 20 (Fri), 8:30 AM – 10:30 AMRoom R316

1. RGMAP: Reinforcement Graph Learning and Momentum Adaptation Prototype Network for EEG Emotion Recognition Weiyang Chen, Yi Zhang, Yuxuan Guo and Weiwei Huang [ID 264]
2. PGD-Mamba: Prior-Guided Dual-domain network with Mamba for Accelerated MRI Reconstruction Lei Wang, Jun Li, Xuan Yang, Jianhua Xu and Bo Yang [ID 294]
3. Heterophilic Hard-Pair Graph Learning for Molecular Property Prediction Yafei Yang, Huifang Ma, Rui Bing, Yuteng Zhang and Meihuizi Jia [ID 416]
4. A Topology-Prior Direction-Decoupled Attention Framework for Trial-Level EEG Emotion Recognition Jiawei Cong and Kai Zhao [ID 527]
5. Learning to Weigh: A Bi-level Reweighting Approach for Cross-Subject EEG Emotion Recognition Shubing Yu, Quanlin Chen, Rui Xiao and Jingcong Li [ID 691]
6. Multimodal Graph Neural Network with Node Representation Decoupling and Distance Regularization for Brain Age Prediction Si Fu, Zhengyuan Lyu, Zhiwei Song and Xiaojuan Guo [ID 730]
7. LeadGLF: Lead-Aware Global-Local Field Learning for ECG Diagnosis and Missing-Lead Reconstruction Cunwei Yu, Xianghong Lin, Xiangwen Wang, Xucheng Li and Wenqian Ma [ID 1344]
8. SNN-DFC: Frame-Wise Dynamic Functional Connectivity Based on Spiking Neural Network Zhao Ganglan, Guo Zhitao, Chen Wanze, Wang Yaokun, Xiao Tiantian, He Yuhao and Long Zhiying [ID 1810]

Session F4.2 (8 papers)Nov 20 (Fri), 8:30 AM – 10:30 AMRoom R319

1. SkinCD: Cross-View Layer-Contrast Decoding for Reliable Skin Disease Classification with MLLMs Ziyu Wang, Xue Wu Zhang, Chang Xu and Fei Qi [ID 334]
2. PocketDiff: Guided Diffusion for Protein Pocket Generation and Optimization Junjie Li, Shifu Luo, Shuquan Su and Jinyan Li [ID 365]
3. Robust Atrial Fibrillation Detection via ECG–Text Alignment and Dual-Teacher Distillation Kai Zhang and Haiying Liu [ID 474]
4. Prototype-Guided Statistical Tokenization and Expert Decoding for Gene Expression Inference from Whole-Slide Images Zikang Liu and Jinglin Du [ID 1256]
5. BioGCC-CaT: Task-Guided Graph-Capsule Evidence Learning for Interpretable Single-Cell Disease-State Mining Ran Yan [ID 1257]
6. DaIG-MIL: Active Lesion Instance Generation for Weakly Supervised Patient-Level Diagnosis in Reflectance Confocal Microscopy Jianghao Yue, Kai Sun, Hongyong Sun, Suiyuan Che, Fangfang Liu, Dingdong Zhou and Wei Huang [ID 1305]
7. SAEDS-MIL: Surrogate-Grid Attention Enhanced Dual-Stream Multiple Instance Learning for Whole Slide Image Classification Jun Wang, Ende Zhang, Minghai Jiao, Yixian Liu, Yinghui Wang, Yi Ren and Chang Liu [ID 1373]
8. Failure-Aware Recovery for Robust Carotid Plaque Segmentation in Ultrasound Videos Xiaoyuan Bao, Yankun Cao, Yijie Lv, Zengxiang Li, Yujue Cheng, Pufang Shan and Zhi Liu [ID 521]

Session F4.3 (6 papers)Nov 20 (Fri), 10:50 AM – 12:20 PMRoom R212

1. Lightweight Boundary-Aware Cross-Modality Attention for Stroke Lesion Segmentation Jingkai Xu, Zhongzheng Lu, Yijie Lv, Zengxiang Li, Pufang Shan and Zhi Liu [ID 794]
2. PAL-Former: Phase-Amplitude Low-Res Transformer for Efficient Medical Image Segmentation Xuewen Chen and Chaoqun Wang [ID 901]
3. LLM-UNet: Augmenting U-Net with Frozen Large Language Models for Continuous Biomedical Signal Denoising Yuhui Cheng and Zeyu Xing [ID 1016]
4. DA-SwinSeg: A Dynamic Adaptation Swin Transformer Framework for Medical Lesion Segmentation Liangqi Ma and Qianyang Peng [ID 1416]
5. DHN-Star: Domain-Adapted Biomedical Semantics and Star Fusion for 3D Multi-Organ Segmentation Haoran Chen, Zhuohang Wang, Tianqi Peng, Yiyang Lin and Haijiang Zhu [ID 1873]
6. SAGF-DTI: Semantic-Anchored Gaussian Fusion of Dynamics-Conditioned Protein Representations for Drug–Target Interaction Prediction Hechuan Jiang, Chonghao Hao, Jianhao Chen, Zhen Xu and Zhaohui Lan [ID 1898]

5. Computer Vision: 3D Vision, Depth and Point Clouds (Sessions F5.1–F5.3, 20 papers)

Session F5.1 (8 papers)Nov 19 (Thu), 8:30 AM – 10:30 AMRoom R321

1. KSG-Net: Key-Sparse and Global-Context Learning for Maritime 3D Ship Detection Zhouyuan Huai, Meiqi Wan, Yan Yang, Minshi Chen, Xin Yuan, Wei Wang and Xiao Wang [ID 227]
2. Bridging Projection and Image Domains: A Two-Stage INR Framework for Ultra-Sparse-View CBCT Reconstruction Jinxia Li, Ziyao Tang, Qinglei Cao and Xiaoqin Tang [ID 859]
3. F-HDRNet: Towards Dynamic Scene Adaptation in High Dynamic Range Imaging Reconstruction Xinlong Bu, Qiang Wang, Hanwei Zhu, Dongyang Zhang and Jieyu Zhan [ID 1274]
4. EfficientADM: Multi-Level Normality Distillation with Variational Feature Reconstruction for Unsupervised Wire Rope Anomaly Detection Zhengyang Chen, Zhengyou Wang, Weipeng Li, Shanna Zhuang, Jing Bai and Congcong Li [ID 1309]
5. Explicable Compressed Sensing Reconstruction via Diffusion Guidance and Error Calibration Yurong Yan, Lijun Zhao and Jinjing Zhang [ID 1319]
6. PSSRoute3D-AD: Hidden-Category Normal Reference Routing for Multi-Category 3D Anomaly Detection Peng Xie, Yu Chen, Mei Zhang and Jinhui Zhu [ID 1594]
7. LineRecover: Mask-Guided Curve Reconstruction for Line Chart Data Extraction under Image Degradation Xiangsheng Zeng, Zeqi Hao, Yiyao Wang, Wei Chen and Di Zhang [ID 1854]
8. OQ-Atoms: Deferred Semantic Commitment for Open-Vocabulary 3D Segmentation Zilei Zhou [ID 450]

Session F5.2 (6 papers)Nov 19 (Thu), 10:50 AM – 12:20 PMRoom R316

1. EventGaussian: Event-Guided Compact Initialization for Lightweight 3D Gaussian Splatting Heng Zhang, Wenchang Li and Yanli Liu [ID 477]
2. GaussianTraj: Scene- and Uncertainty-Aware Trajectory Prediction via Gaussian BEV Splatting Dehai Zhang, Yiran Xie, Ping Wang, Mingzhe Zhang, Zheng Chen and Bowen Chen [ID 510]
3. SFO-SLAM: Semantic-Flow Guided Robust RGB-D SLAM in Dynamic Environments Heng Zhang, Yanlong Guo, Yanli Liu, Zhimin Chen and Neal N. Xiong [ID 1189]
4. Frequency-Guided Cross-Level Feature Enhancement for Few-Shot 3D Point Cloud Semantic Segmentation Zihan Tang, Yong Zhao, Can Zhou, Letian Pei, Xin Si and Jingliang Zhang [ID 1292]
5. SoftWR-Hybrid: Stochastic Score Smoothing for Spatial-Support Diagnostics in 3D Gaussian Splatting Shufeng Li, Junyi Lin, Yiyang Liu, Yun Chen and Muqing Deng [ID 1617]
6. DynInst-GS: Instance-Aware Dynamic Gaussian Splatting SLAM Tan Jia, Li Tan, Xiaojun Zeng and Yixin Peng [ID 1808]

Session F5.3 (6 papers)Nov 19 (Thu), 10:50 AM – 12:20 PMRoom R319

1. Bridging the Semantic-Spatial Gap: Monocular 3D Reasoning in Vision-Language Models 东林 邵 [ID 560]
2. Beyond Best-of-K : Analogical Future Coverage for Multimodal Trajectory Prediction Evaluation Chenkang Li, Shuqiang Gao and Yuzhen Zhang [ID 1162]
3. TwinOcc: Motion-Aware Dual-Stream 3D Occupancy Prediction Aodi Wang, Shuai Cheng, Zuotao Ning, Wei Liu, Yin Zhang and Jun Hu [ID 1374]
4. SAME-Depth: Refining Motion-Aware Cost Volumes for Self-Supervised Multi-Frame Monocular Depth Estimation Jing Wen, Meihong Chen, Guoqi Shu and Minzhi Li [ID 1533]
5. Auditing 3D Shape Retrieval without Supervised Training Using Diffused Geodesic Moments Zhicheng Du, Changyue Liu, Wenji Xi, Zhaotian Xie, Zhuo Deng, Ziheng Zhang, Yang Liu and Lan Ma [ID 1573]
6. Geo2O-HMR: Efficient 3D Human Mesh Recovery via Geometry Consistent Dense One-to-One Supervision Yunpeng Guo, Hui Wang, Feng Xu, Tianpei Zhu and Chunhua Xie [ID 1697]

6. Computer Vision: Detection, Segmentation and Tracking (Sessions F6.1–F6.4, 30 papers)

Session F6.1 (8 papers)Nov 19 (Thu), 1:30 – 3:30 PMRoom R205

1. HDSQ: Heterogeneous Dual-Stream Representation and Dual-Granularity Synergistic Query for Fine-Grained Classroom Action Detection Yinhuang Mao, Jiapeng Gan, Jingyue Li, Yan Liang and Fengbo Zheng [ID 250]
2. MADFNet: Revisiting Marine Animal Segmentation with Depth-Aware Multimodal Fusion Mingkun Zhao, Yuxin Li, Di Wu, Fengyang Liu and Zhibin Yu [ID 331]
3. SE-ADS: Spatially Enhanced Anomaly Detection and Segmentation for Zero-Shot Industrial Visual Inspection Wei Yao, Quanquan Li, Xidong Xi, Conghua Wei and Guitao Cao [ID 476]
4. GeoCueFormer: Geometry-Guided Wavelet Representation and Prediction-Cued Dual-Stage Decoder for Underwater Semantic Segmentation Xian Wu, Xinjin Li, Yiliu Xu, Yining Liu and Yong Jiang [ID 1212]
5. MatchSeg-RL: Scale-Aware Matched Reinforcement Learning for MLLM-Guided Segmentation Zhiyong Xia, Zihao Huang, Ziyu Cheng, Yandan Chen, Weizhao Weng and Ying Gao [ID 1619]
6. DinoCAM: Foundation-Model Guided CAM Refinement for Weakly Supervised Semantic Segmentation Wei Cao and Yong Jiang [ID 1663]
7. DFPC-Lite: Dual-Code Residual Prototype Decoding for Boundary-Sensitive Small-Instance Segmentation Haifeng Fan, Haoran Liu, Jianrong Li, Huizhong Xue and Chuanlei Zhang [ID 1749]
8. DCP: Dual-Chain Progressive Framework with Complementary Supervision for Curvilinear Structure Segmentation Kai Zhu, Demin Shen and Li Chen [ID 1893]

Session F6.2 (8 papers) [Computer Vision: Detection, Segmentation and Tracking + Computer Vision: Document Analysis and OCR]Nov 19 (Thu), 1:30 – 3:30 PMRoom R208

1. GAFTrack:Geometry-First Association for Appearance-Free Online Multi-Object Tracking Heyang Li, Lei Li, Jingyao Zhang, Yucheng Zhang and Shuqin Gao [ID 376]
2. GADFormer: Training-Only Semantic Anchoring for Single-Source Domain Generalization in Driver Behavior Recognition Lei Liang, Yongning He, Tongzhen Si, Jing Wang and Xiaohui Yang [ID 628]
3. SDATrack: Sparse Decoupled Association for Occlusion-Robust Multi-Object Tracking Ke Chen, Chunhua Deng and Junwen Zhang [ID 669]
4. Learning Bit-Plane Semantic Interfaces for CLIP-Aligned Visual Students Xinyi Guo, Haobing Ding, Yuhan Jiao, Zicong Huang, Junxuan Zeng, Chao He, Dongqing Song and Jianyu Xu [ID 899]
5. A Component-Aware and Consistency Graph Learning Approach for Ancient Chinese Character Image Retrieval Tingshan Shao [ID 1314]
6. VADoP: Visual-Aware Damaged Content Prediction for Ancient Documents Minh-Nhut Dang, Thanh-Trieu Duong, Dien Dinh and Long Nguyen [ID 1324]
7. EAD-Spotter: Robust Scene Text Spotting via Environment-Adaptive Decoding and Recognition-Guided Refinement Zhiqiang Wang, Yujia Yin, Sen Teng and Ying Chen [ID 1441]
8. RDC-ROP: Relation-Aware and DAG-Consistent Reading Order Prediction Chengdong Pan, Jun Li and Yi Zhang [ID 1450]

Session F6.3 (8 papers)Nov 19 (Thu), 1:30 – 3:30 PMRoom R210

1. Modality-Preserving and Frequency-Aware Fusion via Semantic Guidance for Infrared-Visible Images Luyao Gao, Mingliang Zhang, Guoxing Wen, Shuang Li and Bin Li [ID 426]
2. Uncertainty-Aware Forest Monitoring: Calibration, Propagation, and Utilization in Degradation Detection and Recovery Prediction Hyeong-Woo Choi [ID 486]
3. CSL-YOLO: An Efficient Object Detection Method Based on Lightweight Multi-Scale Attention for Steel Surface Defect Detection Duo Ma, Yuan Jia, Kaifang Gao, Jundong Tan and Wen Yang [ID 496]
4. Spatial-Frequency Enhanced Detection Transformer for Industrial Surface Defect Detection Taolin Yang, Min Shi, Suqin Wang and Dengming Zhu [ID 706]
5. UFP-Net: Uncertainty-Aware Frequency-Prior Network for Semi-Supervised Change Detection Peicheng Yue, Di Zhang, Qianqian Niu, Yuhang Yan and Huifang Ma [ID 782]
6. HB-DETR: A Frequency-Aware Multi-Scale Transformer with Contextual Bidirectional Feature Fusion for Small Object Detection Xiang Jin, Jing Qi and Yuchuan Huang [ID 1052]
7. Diagnosing and Mitigating Prior-Branch Dependence in Prior-Guided Infrared Small Target Detection Chen Zhang, Jun Jiang, Wenxin Wang, Yujun Zhu and Xin He [ID 1535]
8. SMOT-Net: Mamba-Based Multi-Modal Fusion with Optimal Transport for Off-Road Freespace Detection Zhengkang Gao, Xue Fan, Ruotong Xu, Xiaohui Yang and Weixing Tan [ID 1783]

Session F6.4 (6 papers)Nov 20 (Fri), 10:50 AM – 12:20 PMRoom R316

1. View-Adaptive Adversarial Camouflage for Deceiving Object Detectors Ming Zhang, Cheng Qian, Jie Yang and Ting Bao [ID 643]
2. RASR: Test-Time Adaptation via Decomposing Raw Classification Score for Real-Time Open-Vocabulary Object Detection Xiangyu Gao, Heqian Qiu, Taijin Zhao, Benliu Qiu, Linfeng Xu and Hongliang Li [ID 646]
3. DIC-DETR: Detector-Internal Compensation for Camouflaged Underwater Object Detection Zehua Bao, Boyuan Zhu, Guan Yang, Weisen Zong, Gaoyuan Lu, Chunlei Li and Boyang Qu [ID 662]
4. Revisiting Specific Forgery Features for Generalizable Deepfake Detection Pengfei Lv, Feng Zhang, Bingyu Wu, Haojun Ai and Xinyue Yang [ID 1122]
5. RAD-DETR: Efficient Few-Shot Retrieval-Augmented Adaptation For Incremental Open-Vocabulary Object Detection 颖 张, 芳 罗, 晶晶 郭 and 毅 戴 [ID 1436]
6. SASTCamo: Spatially Adaptive Vehicle Adversarial Camouflage Generation via Natural Texture Selection and Feature-Level Disruption Chenyou Wang, Jun Ren, Bin Qu, Jiang Fang, Hui Zeng and Anjie Peng [ID 1607]

7. Computer Vision: Face, Person and Biometrics (Sessions F7.1, 8 papers)

Session F7.1 (8 papers)Nov 19 (Thu), 3:50 – 5:50 PMRoom R310

1. HiBR: Hierarchical Adaptive Bridge Representation Learning for Multi-spectral Vehicle Re-identification Yuchen Hou, Xitao Zou, Wentao Luo, Yi Xiang and Ying Wu [ID 567]
2. Fast Facial Privacy Protection via Target-Conditioned Distillation Jiawei Liu [ID 657]
3. DRAT-HPE: A Dynamic Region-Adaptive Transformer for Robust Head Pose Estimation Yufeng Huang, Chunhua Deng, Zhenghua Huang, Junwen Zhang and Xing Xu [ID 723]
4. DC-Net: Towards Generalizable Text-Based Person Retrieval via Unified Domain Expansion and Confidence-Aware Alignment Yanbin Hou, Minglei Yang, Xiao Wang and Qi Zheng [ID 781]
5. SPReID: Semantic-Guided Multi-View Proxy Compensation for Occluded Person Re-Identification Chunwang Li, Ruixin Shi and Min Zhi [ID 1304]
6. PEPose: Parameter-Efficient Animal Pose Estimation with Dual Adapter and Adaptive Prompt Lei Li, Yucheng Zhang, Heyang Li, Shuqin Gao, Jingyao Zhang and Jintao Li [ID 1360]
7. Unknown Semantic and Boundary Reliability Learning for Open-Set Video-based Facial Expression Recognition Yuan Yuan, Jinglei Wang, Shumin Li, Wei Xie and Hao Sun [ID 1454]
8. Multi-Granularity Consensus Confidence Selection and Multi-Scale Visual Perception for Text-to-Image Person Re-Identification Jinglei Wang, Yuan Yuan, Shumin Li, Hao Sun and Wei Xie [ID 1504]

8. Computer Vision: Low-level Vision and Image Generation (Sessions F8.1, 8 papers)

Session F8.1 (8 papers)Nov 19 (Thu), 3:50 – 5:50 PMRoom R312

1. Underwater Image Enhancement via Color Prior Guidance and Spatial-Frequency Collaborative Learning Yinghui Song and Lifang Chen [ID 348]
2. Priority-Aware Patch Refinement for Low-Light Image Enhancement Feng Xue, Jiping Zheng and Teng Teng [ID 555]
3. Cloud-Edge Collaboration and Instruction-Driven Mechanism for Multi-Task Adaptive Infrared-Visible Image Fusion Huabin Huang, Renping Xie, Yangyang Wu and Ming Tao [ID 884]
4. S2UIR: Semantic–Structural Prior Unified Modulation for All-in-One Image Restoration Yufei Xu, Hui Wang, Feng Xu, Tianpei Zhu and Guoliang Luo [ID 1055]
5. Qwen-Guided Semantic Conv-Attention Network for Efficient Image Super-Resolution Ruizhi Liu, Hui Wang, Feng Xu, Yulong Wang and Nan Jiang [ID 1138]
6. WT-Fusion: A Frequency-Guided Wavelet-Transformer Network for Multi-Focus Image Fusion Zhehan Gong, Wei Wang, Xiao Wang and Xin Yuan [ID 1159]
7. UDG-Fusion: Uncertainty-Driven Gating between Pixel and Latent Branches for High-Resolution Multi-Focus Image Fusion Beining Wu, Huiming Ding and Zhifeng Xie [ID 1252]
8. PDG-UIE: Underwater Image Enhancement Network via Polynomial Color Correction and Dynamic Gabor Prior Yiting Yang, Jianming Zhang, Yonghang Liu, Gan Cheng and Jun Long [ID 1340]

9. Computer Vision: Medical and Biomedical Imaging (Sessions F9.1–F9.2, 16 papers)

Session F9.1 (8 papers)Nov 19 (Thu), 1:30 – 3:30 PMRoom R312

1. C-SEED: Structured Visual Evidence for Large-Language-Model-Based Citrus Diagnosis Ran Peng, Xiyin Zuo, Yifan Zhang, Keyu Chen and Bowen Zhang [ID 61]
2. Characterizing Text Branch Sensitivity in Medical Vision-Language Segmentation via Evidence Decoupling Ziquan Liu, Zhewei Zhu and Xuyang Shi [ID 277]
3. SpaCon: Spatial Complementary Consistency for Semi-Supervised Medical Image Segmentation Shiqi Shi, Zhengshi Chen and Juanjuan He [ID 509]
4. BECdino: Boundary-Guided Cross-Slice Context Enhancement for Medical Image Segmentation Zhaoxu Lin, Qian Qian and Jiachen Xu [ID 599]
5. TRI-M: Evidence-Aligned Clinical Reasoning Priors for Medical Image Segmentation Songyu Xue, Dangguo Shao, Lei Ma, Sanli Yi and Zongjie Chen [ID 1275]
6. HiCoD-Net: Hierarchical Correlation and Difference Modeling for Video Polyp Segmentation Liuyi Yang, Shao-Chi Pao, Inayatul Haq, Lixin Liang and Bingding Huang [ID 1287]
7. Dual-Pathway nnU-Net for Six-Structure Dental CBCT Segmentation in Implant-Guide Planning Hongde Yang, Tianran Yuan, Yewei Xin, Qian Jiang, Sheng Li and Hua Chen [ID 1633]
8. PTC: Pyramid-Transformer-CNN with Global-to-Local Alignment for Efficient Medical Image Segmentation Jie Shu, Qiyun Ye, Shanlin Liu, Yongmei Zhang, Lijing Tong and Mu-Xing Li [ID 1727]

Session F9.2 (8 papers) [Computer Vision: Face, Person and Biometrics + Computer Vision: Low-level Vision and Image Generation + Computer Vision: Medical and Biomedical Imaging]Nov 19 (Thu), 1:30 – 3:30 PMRoom R316

1. CLT-Net: Connectivity-guided Local Topology Learning for Robust Vessel Segmentation Jiehan Liu and Jiajia Li [ID 552]
2. WaveMorphNet: Frequency-Guided Direction-Aware Retinal Vessel Segmentation Heyi Qiu [ID 692]
3. FAMSA: A Plug-and-Play Frequency-Aware Attention Plugin for Low-Resolution Herb Slice Classification Letian Li, Yuanqi Zheng, Shenning Zhang, Shiyu Tan, Chenghao Miao and Jiamei Jiang [ID 959]
4. NASP-CLIP: NAFAdapter and SAM Feature Fusion for Zero-Shot Medical Anomaly Detection Han Sun and Zhipeng Xu [ID 1118]
5. ReCTNet: A Semi-Supervised Multi-Scale Deformation-Aware Network for Rice Plant Counting and Localization Xinyu Gao, Xiaodong Bai, Yucheng Qian, Xue Zhang and Yu Zeng [ID 1521]
6. TCGMUNet: Topology-Preserving Cross-Frequency Gated Mamba UNet for Histopathological Colorectal Cancer Segmentation Yi Long, Jiayue Li and Lingna Chen [ID 1862]
7. Colour-Bar Inversion for Thermal Facial Expression Recognition Kang Yao, Congcong Tian, Shengjiang Zhang, Ting Dong and Weiwei Fu [ID 1754]
8. Illumination–Texture Disentangled Spatial–Frequency Learning for Shadow Removal Wanzhi Hong, Qingxuan Shi, Fang Yang and Zhenli Zhang [ID 1610]

10. Computer Vision: Remote Sensing and Aerial Imaging (Sessions F10.1–F10.3, 20 papers)

Session F10.1 (6 papers)Nov 20 (Fri), 10:50 AM – 12:20 PMRoom R215

1. LEHS-YOLO: A Collaborative Enhancement and High-Resolution Detection Algorithm for Small Object Detection in Low-Light Aerial Imagery Li Yao, Lingzhi Liu, Yan Wan and Haotian Qi [ID 55]
2. HCF-YOLO: A High-Resolution Cross-Scale Feature Flow Network for UAV Small Object Detection Jundong Tan, Yuan Jia, Lingjing Xu, Kaifang Gao and Duo Ma [ID 513]
3. SDQ-YOLO11: Dynamic Feature Reassembly and Quality-Guided Prediction for UAV Small Object Detection Jipeng Zhou and Gang Shi [ID 896]
4. LHRFE: Lightweight High-Resolution Feature Enhancement for UAV Small Object Detection Jiayan Hong, Yuan Yao, Yalin Wang and Chao Li [ID 985]
5. CSRD-YOLO: Reliable Cross-Scale Feature Learning for UAV Small Object Detection Jiaxin Chen, Zhuo He, Dengyong Zhang, Qi Peng and Yuxu Peng [ID 1247]
6. DPEDet: A Dual Parameter Exposure-correction Detection Network for Enhanced Low-Light Object Detection in Drone Images Qinyuan Jie, Liheng Xu, Jiangyu Chen, Hongwei Gao, Zeming Xu, Jixian Lv and Jian Sun [ID 1846]

Session F10.2 (6 papers)Nov 20 (Fri), 10:50 AM – 12:20 PMRoom R310

1. TCDR-UNetFormer: Progressive Semantic-to-Detail Decoding for Remote Sensing Image Segmentation Ying Cheng and Zili Zhang [ID 472]
2. Dynamic Prototype Alignment with Visual-Textual Priors for Remote Sensing Few-Shot Segmentation Beining Xu, Yu Zhang, Mingliang Zhang, Ying Zhu, Junhan Liu and Bin Li [ID 627]
3. SSCR-Net: Structuralized Spectral Enhancement and Complementary Rectification Network for Remote Sensing Tiny Object Detection Guoqiang Chu, Baolong Li, Gaoshan Song, Song Ming and Qian Sun [ID 832]
4. Multi-Level Directional Structure Mining and Hierarchical Class Interaction for Remote Sensing Image Segmentation Guilin Liu and Liping Ma [ID 1205]
5. DMP-YOLO: A Small Object Detection Method for Remote Sensing Images Ruixiao Wang, Aimin Li, Zhaohui Li and Yujiao Liu [ID 1481]
6. Interpretable Hyperspectral Unmixing Framework with Fixed Endmember Prior and Structured Residual Refinement Ziyi Guan, Jianping Zhang and Qian Liu [ID 1728]

Session F10.3 (8 papers)Nov 19 (Thu), 3:50 – 5:50 PMRoom R321

1. LSAC-Net: A Lightweight Network for Alfalfa Inflorescence Detection in Low-Altitude Multispectral Imagery Junwen Ge, Rui Liu, Yongqi Ge and Zhihua Dang [ID 633]
2. Frequency-Structure Guided Transformer Detection for Small Objects in Low-Altitude UAV Images Lingling Zhang, Yongqing Sun, Jia Su and Xinxiao Li [ID 1093]
3. SpecTr-DETR: A Frequency-Enhanced and Trident-Gated Real-Time Object Detector for Aerial Imagery Zihao Shen and Degang Yang [ID 1117]
4. HALO-RGBT: Asymmetric Receptive-Field Fusion of Visible and Thermal Streams for Tiny Vehicle Detection in Aerial Imagery Haozhe Meng [ID 1149]
5. MSA-DETR: A Multi-Scale Attention-Enhanced Transformer for UAV Small Object Detection Wen Yang, Yuan Jia, Duo Ma, Lengjing Xu and Kaifang Gao [ID 1279]
6. A prediction method for soluble solids content of blueberries based on multi-scale spectral feature fusion and optimal transmission distribution perception Hang Meng, Zhaokui Li, Bingxin Peng, Jiaxu Guo and Yan Wang [ID 1289]
7. EMSF-DETR: An Enhanced Multi-Scale Fusion DETR for Small Object Detection in UAV Images Jiahai Dai, Mingyu Zhao and Hongwei Yang [ID 1412]
8. WingsNet: Efficient Winglet Calibration for RGB–SAR Building Change Detection Tao Liu, Yanbin Liu, Weihua Li, Xiaolin Zhang, Tingyu Wang and Minh Nguyen [ID 1477]

11. Computer Vision: Scene Understanding and Other Vision (Sessions F11.1–F11.4, 30 papers)

Session F11.1 (8 papers)Nov 19 (Thu), 1:30 – 3:30 PMRoom R212

1. SimDA: Data-Efficient Multimodal Alignment via Similarity-Aware Augmentation and Hyperbolic Semantics Zekun Li, Rui Wang, Lihua Jing, Zhishen Wang, Weizhong Zhang and Jinwen Zhong [ID 107]
2. Enroll Once, Identify Everywhere: Few-Shot Cross-Scenario Visual Speaker Recognition via Event-driven SlowFast Networks Junguang Yao, Peichun Hua and Yue Zheng [ID 109]
3. Alfalfa-MMNet: A Multimodal Alfalfa Growth Stage Recognition Network for Low-Altitude UAVs Xiaotong He, Rui Liu and Yongqi Ge [ID 685]
4. Spatial Similarity Field guided Homography Estimation Network Chenyu Xu, Wu Ke and Xilong Liu [ID 803]
5. DSE-Few: Dual-Semantic Support Expansion and Hierarchical Prototype Fusion for Few-Shot Learning Weiran Chen, Zijie Qiu, Huahu Xu and Jue Gao [ID 1197]
6. MIST: A Mamba-Residual Invertible Network for High-Fidelity Image Steganography Chunjing Zhu [ID 1445]
7. SGHPN: Spatially Grounded Hierarchical Prototype Networks for Background-Robust Interpretable Recognition Huimin Lan [ID 1681]
8. Dual View Graph-Enhanced and Adaptive Fusion Network for Multimodal Relation Extraction Han Haolan and Dang Depeng [ID 1804]

Session F11.2 (8 papers)Nov 19 (Thu), 1:30 – 3:30 PMRoom R215

1. AutoConcept: Training-Free Concept-Guided Reranking for Metadata-Available Composed Image Retrieval Tianyu Wang and Tianjiao Wu [ID 247]
2. Confidence-Weighted Decoupled Knowledge Distillation for Image Classification Jinghui Yang, Xianghong Lin and Xiangwen Wang [ID 329]
3. INFUSE: A Training-Free, Constraint-Guided Framework for OCR-MLLM Fusion in Key Information Extraction Bingbing Li, Zhenping Xie, Huihui Shao and Siwei Jiang [ID 886]
4. Iconography-Aware Composed Image Retrieval for Thangka via Selective Vision–Language Verification Hongyi Chu, Fanqi Chu, Han Shu and Le-Minh Nguyen [ID 898]
5. Reliability-Aware Decoupled Stabilization for Time-Varying Micro-Batch Continual Test-Time Adaptation Yijia Rong, Haojie Bai, Yi Luo and Aiguo Chen [ID 1088]
6. DReSS: Atomic Review-Guided Test-Time Optimization for Reliable Multi-Step Garment Image Editing Xiaoxue Wu, Yongmei Zhou, Jianghao Lin, Dong Zhou and Qihua Lao [ID 1516]
7. RouteGraph-Mona: Confusion-Aware Routing Fine-Tuning for Mineral Image Classification Jierui Li, Zhiyuan Qi, Hao Zhu, Yufan Liu, Jixian Liu, Shaojie Jiang, Jianda Wang, Yaqi Liu, Xiaotong Li and Wei Wang [ID 1811]
8. Adapting on the Fly: A Dual-Hedging Framework for Online Learning from Drifting Image Streams Yimin Wen, Xinyue Li, Xiang Liu, Huafu Xu and Bin Deng [ID 1824]

Session F11.3 (6 papers)Nov 20 (Fri), 10:50 AM – 12:20 PMRoom R319

1. CrossDim-DETR: Cross-Dimension Multi-Path Fusion for DETR Yuanwei Li, En Zhu and Li Shen [ID 262]
2. RC-Map: Diffusion-Enhanced Radar-Camera Fusion for Online Vectorized HD Map Construction Fanfan Yang, Tianxiang Cui, Heng Yu and Jiawei Li [ID 701]
3. When Task Gradients Disagree: Dynamic Harmonization for Few-shot Meta-Learning Siyuan Sun and Hongyang Gao [ID 880]
4. Align Where the Words Look: Cross-Attention-Guided Patch Alignment with Contrastive and Transport Regularization for Bengali Captioning Riad Ahmed Anonto, Sardar Md. Saffat Zabin and Mohammad Saifur Rahman [ID 1241]
5. What Backbone Does Drone-Based Geo-Localization Actually Need? Xin Zhou and Xuerong Yang [ID 1251]
6. Dual-Importance Progressive Transmission for 3DGS Liheng Zhang, Weihao Yu and Jin Huang [ID 1347]

Session F11.4 (8 papers)Nov 19 (Thu), 1:30 – 3:30 PMRoom R310

1. Learning Motion-Enhanced and Symmetrically-Consistent Representations for Micro-Expression Recognition Ying Zhu, Mingliang Zhang, Rui Jiang, Ruimin Xu, Shuang Li and Bin Li [ID 419]
2. Robust Physical Adversarial Texture Generation for Ships under Surface Degradations Jingnan Zhang, Xiadong Lyu, Wei Luo, Lancong Guo, Ziqi Liu and Zheng Wu [ID 779]
3. FedRefill: Efficient Federated Semi-Supervised Learning via Global-Guided Filtering and Refilling Jiangcheng Yi and Gang Liu [ID 1006]
4. ReFindVLA: Vision-Language-Action Learning for UAV-Based Dynamic Target Reacquisition Xiaohui Tong, Xinmin Li, Qifeng Wu, Xinyi Zhang and Sitian Wang [ID 1405]
5. PhysiCal: Physical-Consistency Learning for Label-Free Real-Image Metric Adaptation Ziyi Shang and Bo Jiang [ID 1585]
6. Adaptive Rank Allocation for Task Incremental Learning Qian Cheng, Weihu Song, Lei Li, Jian Cheng and Haogang Zhu [ID 1621]
7. MTRGen: Multi-Reward Self-Critical Training and Clinical-Oriented Evaluation for Clinically Accurate Radiology Report Generation Xiang He, Li Chen and Rui Xiong [ID 1823]
8. Where Does JPEG Become Hard to Learn? A Stage-wise Evaluation of JPEG Representations for Visual Recognition Changpei Zha, Tomo Miyazaki and Shinichiro Omachi [ID 1827]

12. Computer Vision: Video, Action and Temporal Analysis (Sessions F12.1–F12.2, 16 papers)

Session F12.1 (8 papers)Nov 19 (Thu), 3:50 – 5:50 PMRoom R212

1. SMGTransformer: A Hybrid ANN--SNN Architecture with Mixed-Gate Cross-Attention for Video Action Recognition Minghao Ma, Sheng-Hua Zhong and Zhijiao Xiao [ID 302]
2. Wold-Net: A Wold Decomposition Framework for Weakly Supervised Video Anomaly Detection Yuheng Cheng, Fengling Hu, Binnan Yan and Tao Zhu [ID 661]
3. Motion-Conditioned Normal Future Distribution Learning for Video Anomaly Detection Adili Maimaiti [ID 891]
4. LM-STI: Learning Modality Spatio-temporal Interaction for Skeleton-based Action Recognition Congcong Li, Caizhi Gu, Shan Zhang and Zhengyou Wang [ID 969]
5. Weakly Supervised Video Anomaly Detection via Difference-Aware Dual Memory and Temporal Dynamics Learning Na Peng, Tianle Ji, Weiwei Zhang and Jun Guo [ID 1031]
6. EP-GCN: Entropy-Guided Attention and Part-Derived Prototypical Learning for Skeleton-Based Action Recognition Zhuang Dong, Qian Huang, Chang Li, Xuejie Zhang and Zimeng He [ID 1236]
7. MCAT-Net: A Multi-Task Causal Adaptive Temporal Network for Micro-Expression Recognition Shuaitao Duan and Shaopeng Wang [ID 1469]
8. Motion-Semantic Dynamic Calibration and Temporal Enhancement Network for Continuous Sign Language Recognition Cuihong Xue, Qingyuan Ma, Tiantian Yuan, Yong Gao and Yan Shi [ID 1793]

Session F12.2 (8 papers) [Computer Vision: Scene Understanding and Other Vision + Computer Vision: Video, Action and Temporal Analysis + Computer Vision: Vision-Language Models and Multimodal Understanding]Nov 19 (Thu), 3:50 – 5:50 PMRoom R215

1. When Events Hurt: Selective Event Trust for Robust Low-Light Enhancement Wentao Wu, Qingchuan Li and Peng Guo [ID 308]
2. Temporally Consistent Tube-Level Token Pruning for Efficient Video Transformers Yongpeng Zhang, Xue-Song Tang and Kuangrong Hao [ID 630]
3. IWE-RAFT: Event-Based Optical Flow Estimation with Multi-Reference Images of Warped Events Chonghao Zhong, Haitao Meng, Wenwei Lin, Gang Chen and Alois Knoll [ID 742]
4. From Relevant Frames to Complete Evidence: Temporal Context Completion for Long VideoQA Jianan Zhang, Xinyi Wei, Wenxuan Liu and Xiaodong Wang [ID 1025]
5. EviTRACE: Temporal Evidence Retrieval for Video-Based Robot Failure Analysis Zhiyan Song, Tao Li, Pingfu Chao and Hong Gao [ID 1423]
6. DeSink: Sink-Residual Visual-Temporal Confidence for Test-Time Video Reasoning Minxuan Xu [ID 1892]
7. UWRA-Net: A Global-Local Collaborative Network with Text Guidance for Urban Waterlogging Risk Assessment Songhui Chen, Qi Tan and Pei Yang [ID 1485]
8. Label Embedding Discrete Online Cross-Modal Hashing Siyuan Huang, Zihao Zhan and Xing Tian [ID 1767]

13. Computer Vision: Vision-Language Models and Multimodal Understanding (Sessions F13.1, 9 papers)

Session F13.1 (9 papers)Nov 19 (Thu), 3:50 – 5:50 PMRoom R316

1. Beyond [CLS] Attention: Rethinking Visual Cues for Effective Token Pruning in VLMs Junhao Liu, Jiaguo Jiang, Ting Jian, Xirui Tian and Guanquan Zhu [ID 143]
2. Continuous Anchor-based Recalibration for Unsupervised VLM Adaptation Yujie Wang, Derui Lyu, Xiangyu Wang and Huanhuan Chen [ID 570]
3. SubMerge: Task-Aware Submodular Coverage for Efficient Visual Token Compression in LVLMs Chuqin Zhou [ID 1170]
4. SLA-VSE: Short-Long Caption Joint Supervision with Adaptive Visual Token Compression for Image-Text Matching Xifan Liu, Renchao Zeng, Siqiang Li and Shouqiang Liu [ID 1191]
5. Rethinking Visual Token Compression in Vision-Language Models: A Cross-Backbone Diagnostic Study of Prompt-Conditioned Visual Measure Matching Xiongwei He, Xiaopeng Zhao, Tianyi Ren, Qi Lu, Ruiyu Zhang and Xinjie Cui [ID 1193]
6. Attention-Guided Patch Selection for Efficient Vision Transformers via Joint Training Jianhua Yang, Yiqiao Cai and Hui Tian [ID 1219]
7. MaDPrune: Medoid-aware Detail-preserving Visual Token Pruning Dachuan Yu, Boyu Diao, Zesen Chen, Zhulin An and Yongjun Xu [ID 1250]
8. TReSM: Text-aware Regional Select-and-Merge for Visual Token Compression in Multimodal Large Language Models You Wu, Yuyang Pan, Zhijian Fan, Jiaming Zhang, Zhike Qiu and Zhiguo Du [ID 1656]
9. TIDAL: Token-Injected Decoder-Aligned Localization for Cross-Task Attacks on Unified VLMs Xuewei Li and Ya Li [ID 1712]

14. Data Mining and Knowledge Discovery (Sessions F14.1–F14.9, 66 papers)

Session F14.1 (8 papers)Nov 20 (Fri), 8:30 AM – 10:30 AMRoom R205

1. Uncertainty-Guided Adaptive Graph Diffusion for Semi-Supervised Tabular Learning Sicheng Che, Peiyu Xu, Changwei Xue, Zheng Li, Yuntao Jia and Guiping Lu [ID 56]
2. FlexStructRAG: Flexible Structure-Aware Multi-Granular Relational Retrieval for RAG Menzhu Chen, Haodong Yang and Jia Cai [ID 125]
3. MAGRec: Multi-modal Alignment Generative Recommendation for AIGC Services Junyuan He, Yu Song, Xuesong Qiu and Ao Xiong [ID 286]
4. Gaussian Knowledge Tracing: Uncertainty-Aware Student Modeling with Short-Term Context Ruoxin Hu, Zhilong Shan and Su Mu [ID 519]
5. EUSR: Evidential Uncertainty-guided Structural Reconstruction for Robust Graph Representation Learning Lin Zhao, Chengjie Shi, Hongchao Wang, Tianming Yang, Dongyang Chen and Ji Xiang [ID 553]
6. Trust-Aware Text-Enhanced Graph Recommendation for Cold-Start and Noisy Content Scenarios Luwei Yin, Juanjuan Deng and Xingyao Yang [ID 1165]
7. CA-GCL: Community-Aware Graph Contrastive Learning with Topology-Preserving Augmentation Kejia He, Bing He, Kaixin Zeng, Shuhua Zhang and Yanhuan Wen [ID 1182]
8. UAFRec: Uncertainty-Aware Adaptive Fusion for Multi-Modal Session-Based Recommendation Wenjie Xia [ID 1495]

Session F14.2 (8 papers)Nov 20 (Fri), 8:30 AM – 10:30 AMRoom R208

1. GSE-xLSTM: Spatiotemporal Decoupled Multivariate Time Series Forecasting with sLSTM and Grouped Spatial MLP Wenxuan Yu, Yujuan Xu, Liming Jiang and Shaomiao Chen [ID 1721]
2. QUEST-SVD: Query-Side Token Compression via Attention-Guided Sink Removal for Efficient Multimodal Late Interaction Huy Hoang Nguyen, Duc Dung Nguyen, Duc Nam Thai, Danh Thai Hoang and Thanh Huong Le [ID 94]
3. High Frequencies Are Useful: High-low Frequency-Decomposition-Enhanced Long-Term Multivariate Time Series Forecasting Macheng Si, Chaojun Li, Qi Zhu, Jie Hao, Qiang Wu and Ran Wang [ID 718]
4. DMRI: Decomposed Multi-scale Fusion and Reliable Interaction for Multivariate Long-Term Time Series Forecasting Hao Feng, Haodi Wang, Libin Jiao, Ye Zhou, Yu Guo and Rongfang Bie [ID 726]
5. Frequency-Aware Time Series Anomaly Detection with Conditional Diffusion Model Li Shen and Weizhu Qian [ID 737]
6. CRIAD: Cross-Scale Residual Reconstruction and Idem-potent Consistency for Time Series Anomaly Detection Jie Lv, Xufei Zhuang, Yuanyuan Zhi, Yupeng Zhao, Qing-Dao-Er-Ji Ren, Xishuang Yang and Ningyu Zhang [ID 844]
7. ProtoTFNet: A Time–Frequency Interaction and Prototype-Constrained Framework for Multivariate Time Series Anomaly Detection Weijun Tang, Lijuan Peng, Liang Wang, Siyuan Liu and Zongjun Ren [ID 1002]
8. TRMDR: Residual-Guided Masked Repair for Multivariate Time-Series Anomaly Detection Yinghao Wu, Minghui Liu, Maihemuti Maimaiti, Tao Wang, Junyi Liang and Zhifan Pan [ID 1270]

Session F14.3 (6 papers)Nov 20 (Fri), 10:50 AM – 12:20 PMRoom R205

1. ACQ: A Deployed Two-Stage Framework for Automated Creative Quota Allocation in Large-Scale Online Advertising Ruizhi Wang, Yu Rong, Kai Liu, Bingjie Li, Qingpeng Cai, Fei Pan and Peng Jiang [ID 127]
2. STEER: Scaling Trajectory Estimation with Event-Enhanced Regime Awareness for Elastic Autoscaling in Cloud-Native Microservices Nannan Zhao, Hao Wang, Jiameng Zhang, Bo Wang, Yachen Shen, Fan Yang, Zhijie Huang, Shujie Han, Xiao Zhang and Xiaonan Zhao [ID 284]
3. VISOR: Observation-Aware Prefix Modeling for Early-Stage Website Fingerprinting Hailong Jiang and Xiuliang Mo [ID 681]
4. Lifecycle-Aware Management of Algorithmically Discovered Quantitative Factors Muyao Zhong and Peng Yang [ID 820]
5. Optimal Multi-way Decision Trees for Stratified Sampling in Online Controlled Experiments Tomoka Takei, Shunnosuke Ikeda and Yuichi Takano [ID 892]
6. Deployment-Realistic Evaluation of Cost-Sensitive Sequential Rankers: A High-Frequency Crypto Benchmark Quy Tran [ID 1870]

Session F14.4 (8 papers)Nov 20 (Fri), 8:30 AM – 10:30 AMRoom R210

1. FinInvest-GTCN: A Graph-Temporal-Causal Network for Risk-Calibrated Venture Capital Prediction Chongli Wu and Yi Zhang [ID 172]
2. A Dynamic Adaptive Energy Prediction Method for Photovoltaic-Powered Sensors in Indoor Abruptly-Changing Light Environments Zhugui Jin, Yongqi Ge, Rui Liu and Guangpu Yin [ID 629]
3. Causality-aware Soil Organic Carbon Prediction under Structural Heterogeneity Guanting Wu and Wei Chen [ID 741]
4. TopoWindGraph: Relation-Gated Multigraph Reasoning for Structured Time Series Prediction Jie Wang, Yongsheng Wang, Dandan Meng, Pei Du, Jiaxuan Lu and Gang Zhu [ID 861]
5. AdaDecompTST: Decomposition with Heterogeneous Encoders and Adaptive Frequency Enhancement Junchao Ma, Jiahao Yang, Manyou Ma, Xiaoxuan Zheng, Naveed Khan, Mustafa A. Al Sibahee and Zaid Ameen Abduljabbar [ID 866]
6. Meta-Multigraph Contrastive Learning for Link Prediction on Heterogeneous Graphs Zi Ye, Yongli Wang, Honghua Huan and Xiaohong Zheng [ID 1099]
7. Learning Microscopic Causal Heterogeneity for Dynamic Graph Link Prediction Yuhong Zhao, Yao Guo and Jingyu Wang [ID 1254]
8. CE-HGCN: Contrastive-Enhanced Hypergraph Convolutional Network for Traffic Accident Risk Prediction Wenxuan Yu, Xinyi Feng, Ning Yang, Liming Jiang and Shaomiao Chen [ID 1695]

Session F14.5 (6 papers)Nov 20 (Fri), 10:50 AM – 12:20 PMRoom R208

1. Censoring-Integrated Fréchet-Bounded Learning for Structurally Valid Multi-Endpoint Cardiovascular Risk Surfaces Xiuze Dong, Kangfei Xiao, Zhiqiang Liang, Shu Liu, Xiaoyi Duan, Yifan He and Zhao Wang [ID 237]
2. Keyphrase Extraction using Interpretable Network Representation Learning Corina Florescu, Avijeet Shil and Wei Jin [ID 370]
3. BaDGC: Towards Robust Deep Clustering via Bidirectional Adversarial Inference and Dynamic Graph Constraints Zhicun Chang, Rui Wang, Jin Zhou, Shiyuan Han, Yingxu Wang, Tao Du, Cheng Yang and Bowen Liu [ID 619]
4. Hypergraph Influence Maximization via Hierarchical Co-evolutionary Deep Reinforcement Learning Yachao Yuan, Pengyao Xu, Jinchao Huang, Bin Yang, Chong Di and Jinbao Li [ID 1310]
5. FISCAL: Sparse Cross-Modal Multi-Agent Learning for Explainable Financial Risk Surveillance Haijie Zhi, Baichen Shi and Zhiguo Tao [ID 1474]
6. CoBo: A Coverage-Boundary Representation for Interpretable At-Risk Prediction from Student Reflections Gen Li, Li Chen, Sijie Xiong, Boxuan Ma, Daisuke Deguchi, Takayoshi Yamashita and Atsushi Shimada [ID 1505]

Session F14.6 (8 papers)Nov 20 (Fri), 8:30 AM – 10:30 AMRoom R212

1. RMR-Sleep: Residual Mutual Reconstruction with Subject Adaptation for Multimodal Sleep Staging Yitao Li, Weihao Pan, Kaikun Huang and Jiahui Pan [ID 274]
2. LLM-based Probabilistic Item Embedding with Dense Self-Distillation for Sequential Recommendation Haoyu Wenren, Ruijie Hou, Yue Wang and Yingming Li [ID 327]
3. Inconsistency-Aware Self-Paced Multi-View Label Distribution Learning Guojie Lian, Yanshan Xiao and Bo Liu [ID 609]
4. HGID: Hybrid Gradient-Matched Incremental Data Distillation for Continual Recommendation Weiyan Liang, Xiongbao Duan, Shuting Gu, Manyou Ma, Naveed Khan and Junchao Ma [ID 802]
5. Private Synthetic Data Generation for Continuous Data via Adaptive Binning and K-Means Koki Kugimiya, Papangkorn Inkeaw and Vorapong Suppakitpaisarn [ID 1413]
6. Context as a Bridge: Test-Time Adaptation for Tabular Data with Feature Missingness Xinyue Liang and Songcan Chen [ID 1434]
7. Learning Sample Reliability for Multi-label Feature Selection With Streaming Labels Kaihao Zhang, Jinglin Liu, Zhenzhen Sun, Hongbo Zhang and Jinghua Liu [ID 1708]
8. A Point-in-Time Audit of Graph-Snapshot Leakage in GNN Stock Selection Kun Zhang and Chunwei Xia [ID 1744]

Session F14.7 (8 papers)Nov 20 (Fri), 8:30 AM – 10:30 AMRoom R215

1. SCCLRec: Multi-Granularity Stochastic Subspace Cross-Layer Contrastive Learning for Graph Collaborative Filtering Rong Qian, Yuhao Zou and Haonan Dong [ID 305]
2. ReCalRec: Collaborative Reliability Calibration of LLM-Derived Semantic Evidence for Graph Recommendation Dahao Jian, Jiemin Chen and Jianguo Li [ID 366]
3. A Zero-shot Generalized Graph Anomaly Detection Framework via Node Reconstruction Phan Nguyen, Dat Cao, Hien Chu and Khue Hoang [ID 412]
4. RACE-CT: Collaborative Transition Evidence Learning with Relation-Aware Calibration for Sparse Next-POI Prediction 宇豪 王, Dengshi Li, Zhiming Zhan, Chun Sun, Xiangdong Zou, Yuxin Li and Hui Li [ID 449]
5. CARES: Calibrated Action-Residual Event Scoring for Efficient Strict-Streaming Evidence in Robot Anomaly Detection Haochen Li, Junhao Wei, Dexing Yao, Yifu Zhao, Yanxiao Li, Zhaoyan Mai, Jiamu Yu, Baili Lu, Sio-Kei Im, Xu Yang and Yapeng Wang [ID 1184]
6. RAGPos-Bench: Diagnosing Evidence Position Bias in Retrieval-Augmented Generation Hantian Zhang, Linghang Zeng and Wentai Wu [ID 1294]
7. Open-Set Intrusion Detection Using Complementary Discriminative-Reconstruction Evidence Yunfeng Wu, Kehao Chen, Jiaqing Li, Wei Dong, Bing Huang and Qingmin Han [ID 1536]
8. GLCRL: Global-Local Collaborative Representation Learning for Graph Anomaly Detection Xueru Cui and Yanjun Lu [ID 1669]

Session F14.8 (8 papers)Nov 20 (Fri), 8:30 AM – 10:30 AMRoom R310

1. DST-KT: Difficulty-Aware Spatiotemporal Knowledge Tracing with Adaptive Regularization Boke Liu and Xiankun Zhang [ID 340]
2. A Dual-Level Knowledge Graph Framework for Vietnamese Legal Document Retrieval Nam Doan Dang Phuong, Kiet Nguyen Anh, Tung Le and Huy Nguyen Tien [ID 582]
3. Relation-aware Hierarchical Reinforcement Learning for Knowledge Graph Recommendation Jiaming Deng, Peiyang Chen, Manyou Ma, Junchao Ma, Mustafa A. Al Sibahee and Zaid Ameen Abduljabbar [ID 752]
4. Candidate-Logit Decoupled Knowledge Distillation for Resource-Constrained Network Intrusion Detection Wangqingjia Qiao, Xinkai Wang, Yaoxuan Wang, Jiacheng Feng and Beibei Li [ID 974]
5. Decoupling Retrieval and Local State for Sparse-History Knowledge Tracing Chenlong Liu, Yang Wu, Rui Hou and Yahong Li [ID 1351]
6. Knowledge-Enhanced Adaptive Graph Learning for Multimodal Electronic Health Records Xiaoxuan Luo, Wei Yu, Hao Wang and Sihao Liao [ID 1527]
7. Temporal Knowledge Graph Completion through Compound Geometric Operations in 3D Space Bao Nguyen and Long Nguyen [ID 1773]
8. Trust the Edge, Doubt the Arrow: Auditing Validation Signals for Text-Mined Clinical Causal Knowledge Graphs Su Lan, Shicheng Zhou, Yiqi Wang, Yanming Zhu, Xuefei Yin and Alan Wee-Chung Liew [ID 1830]

Session F14.9 (6 papers)Nov 20 (Fri), 10:50 AM – 12:20 PMRoom R210

1. TReST: Trend-Guided Residual Spatio-Temporal Modeling for Traffic Forecasting Dapei Chen, Yumeng Guo and Bo Wu [ID 645]
2. SCAN-RI: Integrating Overshooting Top Statistics with Spatiotemporal Cross Attention Network for Tropical Cyclone Rapid Intensification Forecasting Jianhang Chen, Xuanzhi Chen, Guanbo Zhou, Qifeng Qian, Yadong Xue, Zhifeng Fan and Yajing Xu [ID 679]
3. RASF: Protocol-Aware Grounding for Cross-Center Vital-Sign Forecasting Jianyu Xu, Ruiqi He, Zicong Huang, Yuhan Jiao, Chao He and Dongqing Song [ID 1030]
4. SecCWE-Rank: Risk-Calibrated Taxonomy Attention for Temporal CVE-to-CWE Ranking Ya Liu, Yu Heng Li, Zhenkai Ma and Yang Li [ID 1157]
5. PRISM-FSCIL: Boundary-Aware Prototype Learning for Few-Shot Class-Incremental Malicious Traffic Classification Taizhong Dai, Xiaohui Han, Guangqi Liu, Zihan Jia and Chuanhao Jing [ID 1564]
6. MESA: A Framework for Cross-Network Node Classification under Coupled Class Imbalance and Label Shift Xiaoliang Huang and Yuhong Zhang [ID 1897]

15. Decision Theory, Game Theory and Planning (Sessions F15.1–F15.2, 13 papers)

Session F15.1 (6 papers)Nov 19 (Thu), 10:50 AM – 12:20 PMRoom R321

1. ACIR-Race: Locked-Budget Deployment Selection for Low-Data Action-Contact Imitation Yuanhang Ling, Chong Cheng and Yang Lu [ID 21]
2. HCM-MEDA: A CNN-Mamba Surrogate-Assisted Evolutionary Framework for High-Dimensional History Matching Xiangling Meng and Ziming Li [ID 118]
3. Decision-Preserving Opponent Representation in Private-Budget Colonel Blotto Yuanyuan Zhang, Gang Xiao, Feng Ye, Zhipeng Du, Lingtao Xue and Jiaqi Wang [ID 255]
4. Frontier Construction Policy Network for Cardinality-Constrained Portfolio Optimization Zhongcai He [ID 789]
5. D4-SVI: Training-Free Symmetry-View Inference for Fixed PARCO Routing Policies Zhenbo Yan, Min Lu, Qian Liu, Botao Liu and Mengxin Gong [ID 1053]
6. Robust Trajectory Reranking with Trajectory-Conditioned Occupancy World Models Di Deng and Jianli Xiao [ID 1765]

Session F15.2 (7 papers)Nov 20 (Fri), 1:30 – 3:00 PMRoom R321

1. DyRo-MCTS: Enhancing Offline Scheduling Policies with Robust Online Planning in Dynamic Job Shops Ruiqi Chen, Yi Mei, Fangfang Zhang and Mengjie Zhang [ID 480]
2. Mode-Specific Multi-Tree Genetic Programming for Online Scheduling in Heterogeneous Satellite Constellations Ziruo Fang, Ningbiao Tang, Zhencai Zhu and Wen Chen [ID 190]
3. Style-Drive: Diverse and Assignable Driving Styles for End-to-End Planning Ying Xiao and Zhengchao Zhang [ID 224]
4. CG-LR-MMD: Conflict-Graph-Guided Layered Resolution for Multi-Robot Motion Diffusion Yanjun Gao, Yanmiao Jia, Li Su and Lei Zhang [ID 674]
5. Towards Efficient Scheduling for Long-running Applications on Heterogeneous Clusters Shuaixiang Zhang, Hao Tong, Changwu Huang, Ke Tang and Xin Yao [ID 894]
6. RRT-FConnect: Fast Bidirectional Path Planning in Cluttered and Narrow-Passage Environments Truong Quynh Chi, Vu Quoc Huy and Le Hong Trang [ID 1707]
7. AGDiffuP: Adaptive Classifier-Free Guidance Diffusion Planner for Autonomous Driving in Long-Tail Scenarios Hangyu Zhou, Chenkang Li, Hao Wei, Yuzhen Zhang and Pei Lv [ID 1724]

16. Evolutionary Computation and Optimization (Sessions F16.1, 8 papers)

Session F16.1 (8 papers)Nov 19 (Thu), 3:50 – 5:50 PMRoom R319

1. LISA-GP: LLM-Induced Structural Abstraction for Knowledge-Guided Genetic Programming Hyper-Heuristics in Dynamic Cloud Resource Allocation Shuai Zhang, Gang Chen, Hui Ma and Chen Wang [ID 372]
2. A Q-Learning-Enhanced Marine Predator Algorithm for Multi-UAV Path Planning Yuelin Ma, Hao Chen, Zeren Zhang and Guoliang Fan [ID 1019]
3. MoFASE-VLA: Constraint-Aware Multi-Fidelity Search for Frozen Multimodal Policy Adapters Zichen Wei, Ping Zhang, Tingyu Ye, Jiahua Wang and Fang Li [ID 1180]
4. Hierarchical Multi-Agent Reinforcement Learning-Based UAV Scheduling for Integrated Sensing, Communication, and Computation Xiaoqing Li, Weidong Bao, Hui Yan, Zhenxing Xu, Yaqian You and Ji Wang [ID 1367]
5. A Three-Dimensional Code Evolution Framework for Automatic Heuristic Design with Large Language Models Yuming Xiao and Yalan Zhou [ID 1381]
6. EBS-AT : Evolutionary Beam Search Guided Auto-tuning for Deep Learning Compilers Tianping Wu, Haibin Liang, Wenhui Chen, Zihua Liu and Shaojian Qiu [ID 1397]
7. Base-Solver Strength Governs the Value of Learned Configuration: A Controlled Study on the Non-Linear Fixed-Charge Transportation Problem Cosmin Sabo, Petrica Pop and Adrian Petrovan [ID 1660]
8. On Automated Design of VRP Constructive Heuristics David Sedlak, Rostislav Cervenka and Michal Bidlo [ID 1869]

17. Explainable and Responsible AI (Sessions F17.1–F17.2, 16 papers)

Session F17.1 (8 papers)Nov 19 (Thu), 1:30 – 3:30 PMRoom R319

1. A Frequency-Driven Activation Mechanism for Stealthy Trojans in Graph-Based Interconnection Analysis for HPC Systems Jun Zhou [ID 29]
2. Balancing Cost and Robustness for Time Series Counterfactual Explanations Hongnan Ma, Yiwei Shi, Guanxiong Sun, Mengyue Yang and Weiru Liu [ID 528]
3. DUKAN: Disentangled and Uncertainty-Aware Kolmogorov–Arnold Network for Multimodal Sentiment Analysis Mingyu Zhou, Lu Wang, Lisheng Xu and Ning Ma [ID 614]
4. Arguing with Rules and Samples: An Argumentation-Enhanced Self-explaining Variant of AdaBoost Guanxi Li and Jianying Cui [ID 992]
5. CARE: Capability-Aware Region-Adaptive Ensemble for Hallucination Detection Yingqi Zhu and Lei Li [ID 1249]
6. LEMW-Net: Local Enhancement and Trend-Aware Adaptive Wavelet Network for Multivariate Time Series Forecasting Xu Yang, Yuan Feng and Bingru Yin [ID 1492]
7. Grounded Is Not Enough: Auditing Object Overlap and Answer Sensitivity in LVLM Explanations Kangkang Sha and Chuqin Zhou [ID 1747]
8. Confounder-Aware Regional Attribution in Early Modern Chinese Vernacular Fiction Shaozhuo Sun, Jian Wu and Yue Zhang [ID 1816]

Session F17.2 (8 papers)Nov 19 (Thu), 1:30 – 3:30 PMRoom R321

1. Training Trajectories Determine Circuit Removability in Annealable Soft-Prior Transformers Zonglin Yang, Ziming Zhao, Wei Tang, Xunyu Jiang, Yihong Liu, Tailin Chen, Zifu Yu and Jiayu Liu [ID 42]
2. Represented Is Not Computed: A Causal Test of Candidate Algorithmic Intermediates in a Transformer Ishita Darade and Sushrut Thorat [ID 87]
3. Human-AI Value Divergence: Evidence from Rank-Based Comparisons of LLMs and Humans Yang Ma, Song Tong, Bo Wang and Kaiping Peng [ID 636]
4. CHTD: Trustworthy Jailbreak Defense for LLMs by Calibrated Hidden-State Trajectory Detection Jialin Li, Xiongming Zhang, Shaoli Wang, Jingmin Liu and Changguo Guo [ID 773]
5. Beyond Accuracy: Benchmarking Emotional Reasoning in MLLMs with Multi-Metric Evaluation Yiheng Zou and Kentaro Go [ID 885]
6. Cross-Domain Semantic Drift as a Measurable Dimension of Platform Health Information Distortion:A Decoupled Large Language Model Measurement and Agent-Based Trust Dynamics Framework Hongjie Zhang and Xinzhu Pu [ID 953]
7. From Support Samples to Support Evidence: Disentangling Parametric Memory and Retrieved Context in Knowledge-Conflict QA Tao Shi [ID 1281]
8. COFT: Confidence-Guided Online Framework for Fine-Grained SSH Behavior Tracking in Encrypted Tunnels Kai Chen, Guangjie Liu, Jiangtao Zhai, Weiwei Liu and Yuewei Dai [ID 1539]

18. Generative AI and Foundation Models (Sessions F18.1–F18.3, 24 papers)

Session F18.1 (8 papers)Nov 19 (Thu), 3:50 – 5:50 PMRoom R205

1. CEAK: Coulombic Efficiency Alignment with KAN Xinhe Li, Yezeng Chen, Zixu He, Zhuoying Feng, Weichen Dai, Kangyu Yi, Ziang Ji, Shuhong Jiao, Ruiguo Cao and Yi Zhou [ID 175]
2. RAMP: Retrieval-Augmented Multi-source Poster Generation for Medical Popular Science Yu Xia, Guoming Wang, Ling Xiao and Siliang Tang [ID 313]
3. PhysDiT: Data-Efficient Diffusion Transformers with Physics-Inspired Adaptation for 3D Radio Field Generation Xiaoyan Wang, Ting Wang, Fengke Meng, Zhengyan Liao, Yike Gao, Shaopu Wang and Chunlei Wang [ID 336]
4. HDC-Diff: Hierarchical Multi-Condition Diffusion for Controllable Symbolic Piano Generation Shiji Liu and Jian Su [ID 338]
5. Feasible-Domain Closure Constrained Diffusion Model for Urban Electric Vehicle Charging-Demand Scenario Generation Rui Ge, Yaoming Ma, Jiahe Chen, Fusheng Tan and Zihui Pan [ID 575]
6. Z3-Grounded Generative Verification for Chemical Reasoning at Inference-Time Zijie Dai, Yubin Luo and Yi Zhou [ID 929]
7. BudgetVerify: Budgeted Claim Verification for Agriculture-Focused Retrieval-Augmented Generation Xiongwei He, Xinjie Cui, Tianyi Ren, Xiaopeng Zhao and Qi Lu [ID 1216]
8. Adaptive Hierarchical Diffusion for Long-Horizon Planning Fengjuan Mao and Benyuan Yang [ID 1713]

Session F18.2 (8 papers)Nov 19 (Thu), 3:50 – 5:50 PMRoom R208

1. FedGDO: One-Shot Federated Learning via Generative Dual-Key Optimization Xiaorong Guo and Maode Ma [ID 214]
2. PBTR: Taming RMSNorm-Exposed Activation Outliers for Low-Bit LLM Quantization Guo Junling, Chen Qirui, Li Xianyu, Jin Dong, Chen Shuangwu and Yang Jian [ID 249]
3. MQ-XMark: Margin Quality Aware Multibit Watermarking for Text Generated by LLMs Si Han Zhao, De Li and Xun Jin [ID 695]
4. AKLBQ: Asymmetric Bidirectional KL-Based Boundary Optimization for Low-Bit Post-Training Quantization Di Wang, Junzhang Ren and Zhen Zhang [ID 719]
5. InMark: Intrinsic Watermarking for Robust Provenance in Text-to-Image Generation Di Zhang, Xinyu Gao, Huayuan Ye, Junli Fang, Yingjie Hu and Wei Chen [ID 1121]
6. PRSM: Projection-aware Residual with Stable Matching for Retraining-free LLM Pruning Juan Fang, Zexin Wang, Xiaoning Zhai and Shixing Ren [ID 1130]
7. SwiftK: Accelerating LLM Decoding via Query-Aware Dynamic Channel Selection Qunchao Lin, Boyu Diao, Zhaofeng Shen, Siyuan Qu and Yongjun Xu [ID 1240]
8. Toward Fair and Diverse Destination Imagery: Quantifying Destination Stereotypes in Text-to-Image Models Zifeng Liang and Yingpeng Sang [ID 1581]

Session F18.3 (8 papers)Nov 19 (Thu), 3:50 – 5:50 PMRoom R210

1. STAR: SELFIES-based Actor–Critic Transformer for Multi-property Molecular Design Chaoyu Hong, Hang Yuan, Zaiwen Feng, Chen Li and Tadahiko Murata [ID 312]
2. Domain-Independent Dead-End Detection via LLM-Synthesised Predicates Zhendong Du and Kenji Hashimoto [ID 438]
3. RoboHalt: Adaptive Termination for VLA-Based Robot Execution Shuyan Guo, Xin Liu, Qichen Tan and Pingfu Chao [ID 678]
4. Exploring Anisotropy in Soft Prompt Compressors for Effective Context Compression Guojie Liu, Jianing Yang, Yiqi Wang, Zhouyang Jia, Long Peng, Bin Ji and Jie Yu [ID 714]
5. Hydra: Adaptive Multi-Stage Adversarial Attacks on Multimodal Medical RAG Systems Linhan Li, Zui Yin and Zhiguo Tao [ID 760]
6. Low-Tool-Tax Dynamic RAG for Black-Box Scientific Reasoning Xianglong Cao [ID 788]
7. Structure-Guided RAG with Counterfactual Exemplar Synthesis for Radiology Impression Generation Wenjian Ding, Binghua Li, Tong Liang, Xulin Deng, Ryutaro Himeno, Toshiaki Akashi and Zhe Sun [ID 1260]
8. DUET: Dual-Branch User Evidence Transduction for Personalized Large Language Model Reasoning Yimeng Ma, Xiao Zhang and Zheng Ye [ID 1631]

19. Knowledge Representation, Reasoning and AI Foundations (Sessions F19.1–F19.4, 26 papers)

Session F19.1 (8 papers)Nov 19 (Thu), 8:30 AM – 10:30 AMRoom R319

1. Hyper-Relational Knowledge Graph Completion with Negative Candidate Recall and Hierarchical Filtering Keying Na and Shuang Liu [ID 15]
2. DHN: Dual-Channel Graph Pretraining with Hard Negative Mining for Few-Shot DGA Fault Diagnosis Xiaojun Yang, Yu Lei, Wenxuan Chen, Binfeng Yu, Yang Zhang, Dongyun Xu, Shubo Yu, Xinjian Liu, Qiming Liu, Jinguo Shen, Zixuan You, Quanyuan Li and Binwu Zhang [ID 375]
3. RelBall: Relation Ball with Quaternion Rotation for Knowledge Graph Completion Yike Liu, Peijia Xie, Chao He and Huiling Zhu [ID 478]
4. Rank-Based Losses for Knowledge Distillation: Towards Principled Ordinal Supervision Vinh-Dat Chau, Sy-Tan Lam and Trong-Nghia Pham [ID 591]
5. Query-Aware Reasoning Tuning for Few-Shot Document-Level Relation Extraction Hongxia Jin and Xiaowang Zhang [ID 1011]
6. ConvGraphQA: Training-Free Conversational Knowledge Graph Question Answering with Explicit Dialogue Memory and LLM Reranking Lei Zhang, Yuli Chen, Yifan Deng and Jun Liu [ID 1167]
7. Accelerating Adaptive KNN via Inductive Fuzzy Priors and Multi-Metric Graph Attention Bo Li, Lijie Zhou, Hui Yin and Xiaoqin Pan [ID 1437]
8. Fuzzy Incomplete Bipolar Argumentation Framework: Computation, Semantics, and Complexity Tong Tong and Zhizheng Zhang [ID 1818]

Session F19.2 (6 papers)Nov 19 (Thu), 10:50 AM – 12:20 PMRoom R215

1. Label Cuts in Layered Automata: Short Explanations for the Regular Constraint Xinyi Zhu and Zonglin Yang [ID 40]
2. Generative Large Language Models are Better Class-Incremental NER Learners Renjie Hua, Yu Xing, Yunhui Liu, Chuan Ren, Jianhua Zhao and Tieke He [ID 1134]
3. Beyond Row-Level Failures: Structured Coverage for Instruction-Following Evaluation Chenmai Shi and Zijiang Yang [ID 1399]
4. GapHop: Gap-Guided Iterative Retrieval for Multi-Hop Question Answering Congcong Tian, Kang Yao and Fei Wang [ID 1642]
5. Klotski-Bench: A Diagnostic Benchmark for Structured Spatial State-Transition Reasoning in Large Language Models Ziheng Zhong and Teeradaj Racharak [ID 1682]
6. DE-RSED: An Efficient Streaming SED Framework Using Dual-Encoding and Retrieval-Augmented Generation Yuxin Zou, Xiaoqiong Xu, Bingbing Meng, Xi Chen, Lei Wang and Meng Xiao [ID 1789]

Session F19.3 (6 papers)Nov 19 (Thu), 10:50 AM – 12:20 PMRoom R310

1. Conditional Riemannian Adaptive Flow Transport for Cross-Subject EEG Decoding Renwei Meng and Hongyi Zhang [ID 189]
2. C-Former: Cycle-Basis Injection for Transformer Architectures via Discrete Hodge Decomposition Vladimir Ilinov [ID 526]
3. An ASP-based approach to Solving General Stochastic Two-Player Games Yifan He and Michael Thielscher [ID 837]
4. Transparent by Construction: Element Scarcity as a Selection Signal for Online Hyper-Heuristics in Weighted Set Cover Yanxiao Li, Junhao Wei, Yifu Zhao, Baili Lu, Xu Yang and Yapeng Wang [ID 928]
5. Who Moved My Columns? SACT for Robust Tabular Classification under Evolving Schemas Weijie Deng and Zhaoyun Ding [ID 1067]
6. From Long-Term Multimodal Histories to Reliable Evidence: Query-Conditioned Visual Verification and Constrained Selection Xiaotong He, Qi Zhang and Huayu Li [ID 1338]

Session F19.4 (6 papers)Nov 19 (Thu), 10:50 AM – 12:20 PMRoom R312

1. Privacy Risk and Utility Trade-offs in Credal Networks under Tracing Attacks Jiaxi Yin, Guiping Tao, Peng Lin and Yun Zhou [ID 776]
2. PACE-LRPD: Persistence-Adaptive Consolidation of the Endpoint and Local Ranking-Preserving Distillation for Online Continual Learning Li Jiongyang, Wenjin Xu, Qiang Wang and Shuang Wei [ID 997]
3. Future-Gradient-Guided Reuse for Sharpness-Aware Minimization Hongru Duan, Lei Guan and Yongle Chen [ID 1131]
4. Decoupling Structural and Semantic Representations for Self-Supervised Graph Learning Renjie Hua, Kaifang Yao, Yunhui Liu, Chuan Ren, Jianhua Zhao and Tieke He [ID 1132]
5. Semantic Preservation and Local Non-Manipulability in Extension Aggregation Weiwei Chen [ID 1204]
6. ATLAS: Learning Modality-Specific Coordinate Systems for Shared Representation Learning Arij Hraghi, Akram Benamar, Mohamed Djallel Dilmi and Faten Chaieb-Chakchouk [ID 1825]

20. General AI (Sessions F19.5, 6 papers)

Session F19.5 (6 papers) [General AI + Knowledge Representation, Reasoning and AI Foundations]Nov 20 (Fri), 3:20 – 4:35 PMRoom R319

1. SeMoNet: Adaptive Semantic Selection and Motion Modeling for Referring Video Object Segmentation Junshu Shi, Xiaolin Fang, Tongxin Zhu, Juwei Guan, Teqi Wang and Dongdong Gui [ID 26]
2. HyperLEGNN: Hypergraph Label-Element Neural Networks for Low-Resource Chinese Legal Event Detection Jun Zhu, Hanyu Yang and Xin-Yu Xiao [ID 1802]
3. SPRL: Scalable Policy-based Reinforcement Learning for Dynamic Influence Maximization Wenbin Shen, Shiqing Wu, Weihua Li and Quan Bai [ID 1865]
4. Automated Benchmark Construction for LLM Evaluation: A Source–Generator–Validator–Lifecycle Framework Zhuohang Wu, Ruize Zhang, Yu Fu, Yongqi Kang and Yong Zhao [ID 1843]
5. Mode-Correctness Alignment in Verifier-Free Test-Time Scaling: A Convex Diagnostic of Self-Consistency and Diversity Zhurun Zhang [ID 1666]
6. MAGFold: Graph Folding Memory Mechanism for Multi-Agent Systems Xu Zhang, Shaoyong Guo, Yitao Xiao, Qingnan Wang, Yan Sun and Ye Cheng [ID 1092]

21. Large Language Models (Sessions F20.1–F20.6, 44 papers)

Session F20.1 (8 papers)Nov 19 (Thu), 8:30 AM – 10:30 AMRoom R215

1. UtT: Understand then Think Makes Large Language Models More Robust in Mathematical Xin-Yi Liu and Heng-Yang Lu [ID 5]
2. PRIME: A Clinical Pathway-Guided Framework for Evaluating LLM Reasoning through Interactive Multi-turn Consultations Zepeng Lei, Lianshan Sun, Diandong Liu, Yongyi Han and Qianying Yang [ID 833]
3. ME-RPO: Decision-Level Relative Policy Optimization for Multi-Expert Low-Resource Machine Translation Yuxin Zhao, Nier Wu, Yang Liu, Haopeng Wu, Yatu Ji and Shuo Sun [ID 947]
4. The Score Granularity Gap in Black-Box LLM Classification: A Comparative Study of Confidence Constructions Ao Sun, Tian Sun and Jiaxing Geng [ID 1104]
5. DirEAG: Dirichlet Evidence Aggregation for Calibrating Verbalized Confidence in Mathematical Reasoning Haorui Xu, Yuzhou Zhu and Liyuan Gao [ID 1356]
6. Compass-Landmark Decoding for Mitigating Reasoning Loops in Long-Form Language Model Inference Zhurun Zhang [ID 1402]
7. TriageRA-CCF: Source-Side Clinical Confidence and Coverage Signals for Adaptive Rank Budgeting in Medical LLMs Shucan Ji, Yining Huang and Hongliang Guo [ID 1769]
8. ConfExit: Confidence-Guided Dynamic Inference for Real-Time Low-Resource Machine Translation Yuxin Zhao, Nier Wu, Yang Liu, Haopeng Wu, Yatu Ji and Shuo Sun [ID 1903]

Session F20.2 (8 papers)Nov 19 (Thu), 8:30 AM – 10:30 AMRoom R310

1. Making PEFT Work for InstructABSA: From Catastrophic Failure to Effective Recovery Jiacheng Yang, Yucun Lin, Kai Sun and Jianming Wu [ID 80]
2. Toward Discrete Hyperdimensional Computing with Cyclic Groups Laurent Miclet and Gilles Richard [ID 272]
3. CogShell: Jailbreaking LLMs through Transferable Cognitive Shell Memory Ziyang Dong, Jinli Zhang and Zongli Jiang [ID 396]
4. IntraSlice: Structured Pruning with Block-Intra Distribution Transform for LLMs Meng Li, Peisong Wang, Yuantian Shao, Qinghao Hu, Hongjian Fang, Yifan Zhang, Zhihui Wei and Jian Cheng [ID 401]
5. Structural Conversion of Pretrained Transformers: What Training Strategy Reveals About Attention Substitution Xiaolei Han, Jie Wang, Xutai Cui, Fang Wang, Zhaokun Zhu and Fufang Xu [ID 596]
6. What Average Accuracy Conceals: Directional Answer Instability in Quantized LLMs on Agricultural Knowledge Tasks Xinzhan Li, Houkui Zhou, Huimin Yu and Haoji Hu [ID 1237]
7. CaLLMa: Computation Before LLM Decision Making Under Uncertainty Xinzhan Li, Houkui Zhou, Huimin Yu and Haoji Hu [ID 1238]
8. Cultural Failure as a Third Tier of RAG Failure: Taxonomy,Silent Failures, and Empirical Validation in a Low-ResourceLanguage Setting Charibeth Cheng, Ralf Wadee Hernandez, Ma. China Ortiz, Andrea Li Santos and Jann Railey Montalan [ID 1625]

Session F20.3 (6 papers)Nov 19 (Thu), 10:50 AM – 12:20 PMRoom R210

1. LLM-Aided Reward Design via Dynamic Weight Alignment for Reinforcement Learning Zhen Zeng, Haibo Chao, Yanhao He, Che Hou, Wenjie Chen and Ning Tan [ID 440]
2. A PLM-based Disentangled Sparse MoE Framework for Traffic Flow Prediction Hairong Huang and Hong Zhang [ID 594]
3. AutoCue-CAA: Multi-Domain Bias Mitigation with Controlled Association Alignment Yi Xie, Bin Zhu, Yanxia Feng, Ximeng Liu, Haoran Xie, Fu Lee Wang and Yanghui Rao [ID 624]
4. ResumeParsingBench: Importance-Weighted Evaluation of LLMs for Resume Information Extraction Seifeddine Bettaieb, Helmi Messoussi, Yuejun Guo and Artem Antonenko [ID 1151]
5. Execution-Contrastive Supervision Against False Process Positives in Process Reward Models Chen Shenmeng [ID 1595]
6. Language-Guided Upcycling: Controllable Expert Initialization via Neuron Activation Jingwei Song and Zhiqiang Ma [ID 1689]

Session F20.4 (8 papers)Nov 19 (Thu), 8:30 AM – 10:30 AMRoom R312

1. Beyond Drift: Retention-Aware Prototype Neighborhood Replay for Continual VLA Adaptation Zicheng Xu and Yifan Zhang [ID 571]
2. When Difficulty Signals Are Unreliable: Auditing and Gating Sample-Count Adaptive Test-Time Compute Guanyu Liu [ID 860]
3. Step-Level Routing Expands the Oracle Solution Space for Chain-of-Thought Reasoning Guangyuan Qian, Gengrui Zhang, Dawei Feng, Bo Ding and Yuanzhao Zhai [ID 1548]
4. Global Anchor Memory: Preserving Bursty Attention Anchors for KV Cache Compression in Chain-of-Thought Reasoning Enze Shi and Yan Qiao [ID 1593]
5. LLMs Meet Code-Mixing: Do Code-Mixed Inputs Fall Between Their Monolingual-Only Counterparts? Liujian Tang, Shenghao Huang and Shigeki Matsubara [ID 1646]
6. BroadHalo: Budget-Aware Hallucination Detection from LLM Hidden States with Latent Factuality Resonance Zihan Wang and Le Yang [ID 1738]
7. When LLM Judges Say Tie: Equivalence, Boundary Abstention, and Resolution Loss Liang Zhao, Yong Wang and Jiangzhe Chen [ID 1742]
8. Attention-Gradient Complementary Screening (AGCS): A Token-Level Defense Method Against Large Language Models Jailbreak Attacks Yiming Li, Yonghui Zheng, Baoquan Liu and Yihan Si [ID 664]

Session F20.5 (6 papers)Nov 19 (Thu), 10:50 AM – 12:20 PMRoom R212

1. SAGE: Sensitivity-Aware Gradient-Based Efficient Unlearning for Large Language Models Xiaozhou Hou, Siyu Li and Jianfeng Qu [ID 991]
2. Med-CogVLM: A Medical Vision-Language Model with Visual Dependency Reinforcement Learning Zeyu Lv, Peng Zhang, Qianwen Zhao, Hongxin Dong, Pinle Qin and Fengbo Xie [ID 1163]
3. STATBench: A Chinese Benchmark for Evaluating Large Language Models in Statistical Theory and Practice Ruolin Chen, Fangwei Lin and Bin Wu [ID 1616]
4. Enhancing Portuguese Large Language Models via Curated Data Mixtures and MoE Upcycling Menghao Li and Shaolin Zhu [ID 1632]
5. BCC: Exposing Backdoors in Large Language Models via Bidirectional Causal Consistency Jianwen Fu, Tiankui Zhang and Hongyang Han [ID 1820]
6. CausalHD: Causal Mediation Analysis for Hallucination Diagnosis and Correction in Vision-Language Models Jianbo Liu, Yingjia Sun and Zhihao Wang [ID 1841]

Session F20.6 (8 papers)Nov 19 (Thu), 8:30 AM – 10:30 AMRoom R316

1. An LLM-Powered Two-Step Framework for Classical Chinese Poetry Translation Yihan Wang, Siqi Huang, Ziyu Ye and Tianai Li [ID 795]
2. VisFact: Verification-Oriented Visual Claim Factorization for Faithful VQA Chaodong Tong, Jiayang Gao, Qi Zhang, Nannan Sun, Yanbing Liu and Lei Jiang [ID 925]
3. ZET: A Single-Model Zero-Shot Detector for LLM-Generated Text Zhenyi Qi, Wanwei Liu, Yaohua Wang, Ji Wang, Xinyu Sun, Jianing Chen, Yang Tan, Xiahui Zhu and Haoang Chi [ID 1178]
4. RBD-SC: Recursive Binary Decomposition for Zero-shot Reasoning with Self-Consistency Iok Tong Lei, Han Yu, Yige Yao and Zhidong Deng [ID 1429]
5. FedGL-LoRA: Structure-Aware Federated LoRA for Legal Long-Document Classification Longhuan Wang and Weiye Sun [ID 1446]
6. ReviewSense: Cluster-Guided Evidence Selection for Actionable Review-to-Advice Generation Siddhartha Krothapalli, Kartikey Singh Bhandari, Tridib Kumar Das, Praveen Kumar, Naveen Suravarpu and Pratik Narang [ID 1717]
7. SemFaith: Diagnosing Semantic Faithfulness of LLM-Generated Classical-to-Modern Chinese Restatements Jiangping Huang, Weidong Yuan, Junchen Xiao and Qixuan Gu [ID 1835]
8. Beyond End-to-End Generation: Diagnosing the Node--Topology Gap in Executable Flowchart Reconstruction Siyang Li [ID 1900]

22. Machine Learning and Deep Learning (Sessions F21.1–F21.9, 54 papers)

Session F21.1 (6 papers)Nov 20 (Fri), 1:30 – 3:00 PMRoom R205

1. Dual-Stream Features Enhanced Large Language Model Reprogramming for Bearing Remaining Useful Life Prediction Xiaoyi Duan, Mingxiao Lu, Xiaofeng Hu and Kai Wang [ID 43]
2. UCA-GAN: Weakly Supervised Anomaly Detection under Partially-Observed Data Contamination Peihang Li, Ruofan Yang and Qinliang Su [ID 817]
3. Universal Semantic Enhancement for Weakly Supervised Anomaly Detection on Tabular Data Siyu Zhou, Yuqin Xie, Hao Yuan and Yingjie Zhou [ID 1009]
4. Multivariate Time Series Anomaly Detection Method Based on Differential Dual-Stream and Multi-Scale Graph Attention Huilin Liu, Jing Wang and Jianli Ding [ID 1380]
5. Modality-Preserving Discrete Token Fusion for Self-Supervised Well Log Interpretation Yaobin Wang, Ruiting Sun, Guanwen Zhang and Wei Zhou [ID 1651]
6. BEAM: Preserving EEG Identity Privacy via Block Exchange and Adaptive Mixup Yuxi Dong, Yimin Wen, Weilong Zheng, Chuangquan Chen and Baoliang Lu [ID 1788]

Session F21.2 (6 papers)Nov 20 (Fri), 1:30 – 3:00 PMRoom R208

1. Domain-Aware Inconsistency Interaction Network for Multimodal Fake News Detection Yi Zhang [ID 1858]
2. DPMA: Dual-Phase MILP Acceleration via Weighted Subgraph Partitioning and Coefficient-Sensitive Variable Fixing for Million-Scale Problems Jieyang Xu, Shuli Zeng, Sijia Zhang, Wantong Xie, Feng Wu and Xiang-Yang Li [ID 163]
3. Memory-Guided Fault Tolerance Analysis for Multi-Tenant FPGA CNN Accelerators Yuyao Wang and Yi Zhang [ID 171]
4. AMT-Classifier: Adaptive Multimodal Transformer for Encrypted Traffic Classification Chen Wei and Chunlei Zhao [ID 178]
5. Text-Assisted Non-Textual Denoising for Multimodal Sentiment Analysis Wenqing Wang, Peng Zhang, Shouzhen Shi, Wei Zhao, Fuqiang Wang and Xiaoming Wu [ID 503]
6. Machine Learning-based Two-Stage Graph Sparsification for the Travelling Salesman Problem Bo-Cheng Lin, Yi Mei and Mengjie Zhang [ID 523]

Session F21.3 (6 papers)Nov 20 (Fri), 1:30 – 3:00 PMRoom R210

1. Robust Multimodal Learning under Missing Modalities via Dual-Hypergraph Driven Prompting Weijie Chen, Chuiyi Chen, Zihao Lin and Junguang Su [ID 1048]
2. PSR-CDCNet: Perturbation-Stable Reliability Fusion for Lightweight Speech-Language Multimodal Sentiment Analysis Xingtao Liao and Yong Huang [ID 1213]
3. Prompt-Guided Invariant Disentanglement Framework for Incomplete Multimodal Sentiment Analysis Yuhuan Wei, Yuxia Lin, Zequn Li and Sijie Mai [ID 1291]
4. Prototype Alignment and Adaptive Consistency Regularization for Learning from Label Proportions Xiangbo Liang and Jing Chai [ID 199]
5. Constraint-Aware Reconstruction-Feedback Adversarial Attack for Trajectory Prediction Zixuan Wang, Zhao Ma and Jing Zhu [ID 212]
6. Unsupervised Graph Partitioning for Learning Structural Priors in SAT Solving Xin Chen, Ruowei Ma and Xiutao Feng [ID 413]

Session F21.4 (6 papers)Nov 20 (Fri), 1:30 – 3:00 PMRoom R212

1. Graph-Regularized Stock Return Prediction: Constraints Beat Complex Graphs Guang Yang [ID 544]
2. Semi-Supervised Multi-Label Learning from Single Positive Labels Tian Zhang, Zeqian Zhang, Zhongnian Li, Tongfeng Sun and Xinzheng Xu [ID 578]
3. BPiKoNet: A Bezier-Parametrized and Physics-Informed KAN-DeepONet for Unsupervised Structured Grid Generation Jianpeng Zhou, Weiting Liu, Hongfei Ye and Jianjun Chen [ID 1409]
4. Constructive Scheduling for Temporal Constraints with Deep Reinforcement Learning Marvin Brune, Elias Theis, Constantin Waubert de Puiseau, Hasan Tercan and Tobias Meisen [ID 1771]
5. Confidence-Calibrated Contrastive Predict-and-Search with Feasibility Refinement for MILPs Jinyi Zhong, Sijia Zhang, Yuke Wang, Shuli Zeng, Feng Wu and Xiang-Yang Li [ID 236]
6. Fairness-Oriented Testing through Pairwise Retraining Jingran Yang, Muxiang Zhang, Lingfeng Zhang and Min Zhang [ID 1059]

Session F21.5 (6 papers)Nov 20 (Fri), 1:30 – 3:00 PMRoom R215

1. Boundary Failure in AI-Based SAM Diagnosis: A Clinically-Informed Evaluation Framework Under Realistic Missingness Franck-Arnaud Liehouan, Masato Kikuchi and Tadachika Ozono [ID 1468]
2. LSC-SCA-EG: Calibrated Selective Diarization Correction for Robust Speaker-Attributed ASR Chen Fu, Hongbin Liang and Lilin Sui [ID 1524]
3. Dual-Objective Temporal Rank-Calibrated Hazard Boosting for Budgeted Corporate Violation Surveillance Zheyao Sun, Pengyu Lu, Qi'An Hu, Xinlan Lou, Taiyu Yang and Yuxiao Jiang [ID 1678]
4. Cost-Aware Self-Auditing for Multi-Hop RAG via Calibrated Information Value Chen Shenmeng [ID 1782]
5. MSTResCaps: A Lightweight Multi-Scale Temporal Residual Capsule Network for EEG-Based User Identification Jiyuan Li, Weilai Guo, Yuming Liao, Qingyu Zeng, Chang Liu, Yingzhe Tong, Zuyu Gu and Songhui Guo [ID 469]
6. TCSP: A Mesoscopic Plasticity Mechanism for Spiking Neural Networks Inspired by Memory Engram Yige Yuan, Jing Zhang, Wenjin Xie and Daiyang Shan [ID 573]

Session F21.6 (6 papers)Nov 20 (Fri), 1:30 – 3:00 PMRoom R310

1. CaPhy-Net: Context-Aware and Physics-Guided Bidirectional Traffic Flow Forecasting for Complex Transportation Hub Services Yibing Lin, Binbin Feng, Kaiyu Zhang and Zhijun Ding [ID 654]
2. TCBiA: Target-Conditioned Bidirectional Attention Network for Open-Set Underwater Ship Verification Shuangcheng Yang, Hao Zhou and Dan Zou [ID 774]
3. Temporal Adaptive Cross-Modal Evidence Retrieval and Query-Based Pooling for Depression Severity Estimation Shiying Huang, Mei Song, Yi Zhu and Li Zhao [ID 1300]
4. Reliability-Gated Dual-Evidence Learning for Lightweight Speaker-Verified Wake Activation Chen Fu, Hongbin Liang, Lilin Sui, Hongbo Huang and Yuqi Liu [ID 1488]
5. Learning Instance-Specific Reduction Policies for the Minimum Hitting Set Problem Laiwen Luo, Zhengling Yangli and Hua Jiang [ID 495]
6. LaSER: A Server-Side Defense Against Delayed-Activation Backdoors Triggered by Federated Unlearning Congwei Luan, Yaowei Zhu, Jixin Ma, Yihong Li and Xiaoyi Zhou [ID 547]

Session F21.7 (6 papers)Nov 20 (Fri), 1:30 – 3:00 PMRoom R316

1. Intention-Enhanced Reinforcement Learning via a Teacher-Student Paradigm for Autonomous Driving Decision-Making Tingting Zhao, Zhenxing Shi, Huijuan Zhang, Chengyue Cao, Jianzheng Liu and Zhengshu Zhou [ID 579]
2. The Trigger You See Is Not the Trigger You Keep: Joint-Trigger Backdoors Can Survive Single-Trigger Defense Enze Shao, Yi Yang, Weihua Li, Yanbin Liu and Quan Bai [ID 650]
3. NSACR: Neuron-Level Subspace Analysis and Client Rescue – A Robust Defense for Federated Learning Xiaotian Liu, Bencan Gong and Shiheng Guan [ID 658]
4. FedAlign-FHE: Alignment-Enhanced Encrypted Federated Learning for Smart Healthcare Hang Zhou, Aiting Yao, Di Shao, Xingrao Ma and Chengzu Dong [ID 698]
5. FUAPS: A Federated Unlearning Method via Adaptive Parameter Selection and Selective Negated Pseudo-Gradients Yuxiao Ding, Hui Xia, Shuo Xu, Le Li and Wenhui Liu [ID 731]
6. FedCVESA: Taking Away Training Data in Federated Learning via Correlation Value Encoding and Segmented Aggregation Chongkai Li, Bang Zhang and Wenjian Luo [ID 1288]

Session F21.8 (6 papers)Nov 20 (Fri), 1:30 – 3:00 PMRoom R319

1. UGT-DSAC: Uncertainty-Guided Replay and Transformer Encoding for LiDAR-Based Mobile Robot Navigation Huiyan Han, Jiepeng Wei, Luyang Li, Liqun Kuang, Rong Zhao, Huaqi Liu and Shengjie Wu [ID 505]
2. Aircraft Taxi Time Prediction with Uncertainty Quantification via a Hybrid GNN-Fuzzy Model Minnan Piao, Weiling Wang, Haifeng Li, Xiaolei Wang, Yufeng Gao and Zhibin Gao [ID 598]
3. Probabilistic Cardiac Image Registration via Function-Space Variational Inference using Task-Specific Priors Yiming Dai, Yong Hua and Xuan Yang [ID 911]
4. Improving Neural Ant Colony System via representation consistent solution scoring for Vehicle Routing Problems Qian Liu, Min Lu, Zhenbo Yan, Zeyu Liu and Zhihui Kong [ID 1125]
5. Singing-Voice-Guardian: Robust and Imperceptible Singing Voice Protection via Latent Diffusion Models Xinyi Wan, Ying Wang, Yuchuan Luo, Zhenyu Qiu, Lin Liu, Jun Zhang and Shaojing Fu [ID 1327]
6. ParetoPath: Preference-Guided Intra-Operator Parallel Pattern Generation for Distributed Transformer Training Xialin Su, Wei Shang, Peng Liang, Mingxiang Zhu, Yanqi Shi and Kaixi Yang [ID 1542]

Session F21.9 (6 papers)Nov 20 (Fri), 10:50 AM – 12:20 PMRoom R321

1. Risk-Aware Static-Dynamic Coupled Policy Representation for UAV Navigation in Mixed Obstacle Environments Jianjie Yang, Xianchang Wang and Runa Song [ID 1567]
2. A spatio-temporal decoupled trajectory prediction network with kinematic priors Yuanyuan Zhi, Xufei Zhuang, Jie Lv, Yupeng Zhao, Qing-Dao-Er-Ji Ren, Xishuang Yang and Ningyu Zhang [ID 812]
3. Latency-Aware Reinforcement Calibrator for Sequential Probabilistic Prediction Yadong Xue, Zekai Pan, Jianhang Chen and Yajing Xu [ID 834]
4. STeSA：Stealth-aware Traffic Sensor Assessment for Diagnosing Spatiotemporal Forecasting Robustness Yuan Gao, Xin Zhou, Junjie Zhou, Xutong Zhou and Yanyan Han [ID 857]
5. FARFuse: Frequency-Aware Aggregation–Redistribution Network for Infrared–Visible Image Fusion Haochen Xie and Yong Jiang [ID 1206]
6. STPF: A Spatio-Temporal Progressive Fusion Network for Traffic Flow Forecasting Yuzhuo Yuan, Yu Feng, Ping Wang, Daohui Ge and Zhaofeng Niu [ID 1398]

23. Natural Language Processing and Multimodal Interaction (Sessions F22.1–F22.3, 13 papers)

Session F22.1 (5 papers)Nov 20 (Fri), 3:20 – 4:35 PMRoom R212

1. TARF: Text-Aware Reliable Fusion for Multimodal Relation Extraction Yachuan Zhang and Yi Guo [ID 222]
2. ThermFi: Cross-Domain Human Action Recognition via WiFi and Thermal Fusion Zhiyi Wu, Jinchao Hu, Siyu Teng, Jiahe Lin and Chao Ma [ID 350]
3. IFNet: Hierarchical Inhibitory Fusion for Robust Multimodal Sentiment Analysis Ningping Li, Ruibao Zhang, Hao Zhang and Yi Zhou [ID 354]
4. Reliability-Calibrated Syntactic Fusion for Aspect Sentiment Triplet Extraction Yongliang Ye and Biqing Zeng [ID 592]
5. RDFNet: Reliability-aware Dynamic Fusion Network for Multimodal Sentiment Analysis Jianxin Tang and Biqing Zeng [ID 768]

Session F22.2 (4 papers)Nov 20 (Fri), 3:20 – 4:35 PMRoom R215

1. Hierarchical Reliability-Aware Sparse Latent Routing Fusion for Multimodal Intent Recognition Qing Qi, Xianyong Li, Dong Huang, Yanli Lee and Yajun Du [ID 1498]
2. MACO: Masked Query Learning with Cosine-Aware Orthogonal Decoupling for Multimodal Sentiment Analysis Shouzhen Shi, Peng Zhang, Xiuli Du, Wei Zhao, Fuqiang Wang and Xiaoming Wu [ID 1554]
3. SDMR: Self-Distilled Missing-Aware Reconstruction for Robust Multimodal Sentiment Analysis Jing Yang and Xia Li [ID 1768]
4. SelfAnchorAlign: Anchor-Mediated Vision-Language Alignment without Negative Sampling Zhenyu Liu and Jinqing Qi [ID 315]

Session F22.3 (4 papers)Nov 20 (Fri), 3:20 – 4:35 PMRoom R310

1. Safeguarding Large Language Models Against Jailbreak Attacks via In-Decoding Safety-Awareness Probing Yinzhi Zhao, Ming Wang, Shi Feng, Xiaocui Yang, Daling Wang, Yifei Zhang and Sisi Wang [ID 1357]
2. A Simple Language and Multimodal Debiasing Method for Robust Visual Question Answering Yihan Zhao and Yuncheng Jiang [ID 1703]
3. SafeSight: Auditing False Alarms in Embodied Hazard Perception Wang Binghao and Rafal Rzepka [ID 1736]
4. PU-SeD: Serialization-Shared Dual-Branch Encoder for Point Cloud Upsampling Yicheng Qian, Lin Qiao, Xiaochuan Wang and Ruijun Liu [ID 1878]

24. Machine Learning and Deep Learning (Sessions F22.4, 4 papers)

Session F22.4 (4 papers) [Machine Learning and Deep Learning + Natural Language Processing and Multimodal Interaction]Nov 20 (Fri), 3:20 – 4:35 PMRoom R321

1. Entity-Grounded Dual-Repository Reasoning for Multi-Hop Question Answering Mingyu Chen and Ruifang Liu [ID 1880]
2. CoRformer: A Residual Feature Reuse Model for Long Sequence Time Series Forecasting Bingru Yin, Yuan Feng and Xu Yang [ID 1561]
3. Predicting Predictability: An investigation of LSTM model performance for the forecasting of chaotic time series Kevin Nolan, Mark McCartney, David H. Glass and Samuel Moore [ID 1667]
4. FreDEA: Frequency-Aware Dual External Attention for Time Series Forecasting Hongmin Qin, Yilong Li, Taohui Huang, Liming Jiang and Huanyu Wang [ID 1698]