|
顶会论文交流 |
|
顶会论文交流环节拟邀请过去一年在 ICML、NeurIPS、COLT、KDD、IJCAI、AAAI、CVPR、ICCV
等顶级国际会议上发表论文的作者前来交流。交流形式为
“Spotlight + Poster”
。作者首先就论文内容在会上做简短报告，时间为2分钟（120秒），之后在墙展报告环节与参会者进行面对面深入交流。值得一提的是，简短报告最好具有一定的吸引性，那样有助于吸引更多参会者通过墙展报告环节了解您的研究工作。 如果您希望得到成果展示的机会，请按照此格式提交申请表。该申请表将由专家进行评审，通过评审的申请人将获得 Spotlight 展出的机会。
|
Poster Spotlight Session (I)
| 序号 | 报告人 | 报告人单位 | 论文题目 | 会议名称 |
| 1 | 周植 | 南京大学 | A Theoretical Study on Bridging Internal Probability and Self-Consistency for LLM Reasoning | NeurIPS 2025 |
| 2 | 陶小玉 | 中国科学技术大学 | InstructTime: Advancing Time Series Classification with Multimodal Language Modeling | WSDM 2025 |
| 3 | 付嘉乐 | 东南大学 | Fast Large Language Model Collaborative Decoding via Spectulation | ICML 2025 |
| 4 | 赵丽君 | 山东大学 | Attraction Diminishing and Distributing for Few-Shot Class-Incremental Learning | CVPR 2025 |
| 5 | 黄致远 | 北京邮电大学 | Exploring a Principled Framework for Deep Subspace Clustering | ICLR 2025 |
| 6 | 胡文超 | 南京大学 | Efficient Rectification of Neuro-Symbolic Reasoning Inconsistencies by Abductive Reflection. | AAAI 2025 |
| 7 | 赵张弛 | 西安交通大学 | Improving Memory Efficiency for Training KANs via Meta Learning | ICML 2025 |
| 8 | 李玥颖 | 浙江大学 | A Timestep-Adaptive Frequency Enhancement Framework for Diffusion-based Image Super-Resolution | IJCAI 2025 |
| 9 | 林约拿 | 北京工业大学 | Mitigating Local Cohesion and Global Sparseness in Graph Contrastive Learning with Fuzzy Boundaries | ICML 2025 |
| 10 | 黄文柯 | 武汉大学 | Be Confident: Uncovering Overfitting in MLLM Multi-Task Tuning | ICML 2025 |
| 11 | 李绍园 | 南京航空航天大学 | DM-POSA: Enhancing Open-World Test-Time Adaptation with Dual-Mode Matching and Prompt-Based Open Set Adaptation | IJCAI 2025 |
| 12 | 叶鸣轩 | 中国科学技术大学 | Dynamic Configuration for Cutting Plane Separators via Reinforcement Learning on Incremental Graph | NeurIPS 2025 |
| 13 | 郭平 | 香港城市大学 | MOS-Attack: A Scalable Multi-objective Adversarial Attack Framework | CVPR 2025 |
| 14 | 王俊杰 | 东南大学 | Weakly-Supervised Contrastive Learning for Imprecise Class Labels | ICML 2025 |
| 15 | 朱学峰 | 江南大学 | Collaborating Vision, Depth, and Thermal Signals for Multi-Modal Tracking: Dataset and Algorithm | NeurIPS 2025 |
| 16 | 李巍 | 清华大学 | ZeroFlow: Overcoming Catastrophic Forgetting is Easier than You Think | ICML 2025 |
| 17 | 叶航廷 | 吉林大学 | LLM Meeting Decision Trees on Tabular Data | NeurIPS 2025 |
| 18 | 郑洋 | 电子科技大学 | Integrating intermediate layer optimization and projected gradient descent for solving inverse problems with diffusion models | ICML 2025 |
| 19 | 高恩豪 | 南京大学 | Discovering Symbolic Partial Differential Equation by Abductive Learning | NeurIPS 2025 |
| 20 | 刘铸 | 大连理工大学 | Moreau Envelope for Nonconvex Bi-Level Optimization: A Single-loop and Hessian-free Solution Strategy | ICML 2024 |
| 21 | 王加鑫 | 海南大学 | Hierarchical Shortest-Path Graph Kernel Network | NeurIPS 2025 |
| 22 | 成子健 | 南京大学 | TabFSBench: Tabular Benchmark for Feature Shifts in Open Environments | ICML 2025 |
| 23 | 姚亮 | 河海大学 | RemoteSAM: Towards Segment Anything for Earth Observation | MM 2025 |
| 24 | 孙昊 | 山东大学 | From Pretraining to Pathology: How Noise Leads to Catastrophic Inheritance in Medical Models | NeurIPS 2025 |
Poster Spotlight Session (II)
| 序号 | 报告人 | 报告人单位 | 论文题目 | 会议名称 |
| 25 | 魏来 | 上海交通大学 | First SFT, Second RL, Third UPT: Continual Improving Multi-Modal LLM Reasoning via Unsupervised Post-Training | NeurIPS 2025 |
| 26 | 胡益豪 | 东南大学 | Selective Label Enhancement Learning for Test-Time Adaptation | ICLR 2025 |
| 27 | 何浩源 | 南京大学 | A learnability analysis on neuro-symbolic learning | NeurIPS 2025 |
| 28 | 魏孟 | 中国矿业大学 | Learning from True-False Labels via Multi-modal Prompt Retrieving | ICML 2025 |
| 29 | 李泉江 | 国防科技大学 | Theory-Driven Label-Specific Representation for Incomplete Multi-View Multi-Label Learning | NeurIPS 2025 |
| 30 | 蒋超 | 伯明翰大学 | Trading off quality and uncertainty through multi-objective optimization in batch Baysian optimization | AAAI 2025 |
| 31 | 庄严 | 中国科学技术大学 | Position: AI Evaluation Should Learn from How We Test Humans | ICML 2025 |
| 32 | 李修成 | 哈尔滨工业大学（深圳） | Boundary-Value PDEs Meet Higher-Order Differential Topology-aware GNNs | NeurIPS 2025 |
| 33 | 罗茂林 | 东南大学 | LADA: Scalable Label-Specific CLIP Adapter for Continual Learning | ICML 2025 |
| 34 | 马志远 | 中国科学技术大学 | Advancing Tool-Augmented Large Language Models via Meta-Verification and Reflection Learning | KDD 2025 |
| 35 | 吕立 | 山西大学 | View-Association-Guided Dynamic Multi-View Classification | IJCAI 2025 |
| 36 | 刘伯恒 | 北京理工大学 | Cognitive Predictive Processing: A Human-inspired Framework for Adaptive Exploration in Open-World Reinforcement Learning | NeurIPS 2025 |
| 37 | 康玻瑞 | 南京大学 | Dynamic Multi-Layer Null Space Projection for Vision-Language Continual Learning | ICCV 2025 |
| 38 | 马佳明 | 中国科学技术大学 | BiST: A Lightweight and Efficient Bi-directional Model for Spatiotemporal Prediction | VLDB 2025 |
