Poster Session
4月18日（周六）15:30 – 17:00 · 共 50 篇
P01
A Convergence Analysis of Adaptive Optimizers under Floating-point Quantization
李继初（中国人民大学）
P02
AdaDetectGPT：Adaptive detection of LLM generated text with statistical guarantees
周弘毅（清华大学）
P03
Adam Reduces a Unique Form of Sharpness: Theoretical Insights Near the Minimizer Manifold
李星汉（清华大学）
P04
Adaptive Discretization for Consistency Models
白家宇（华中科技大学）
P05
Advancing Spatiotemporal Representations in Spiking Neural Networks via Parametric Invertible Transformation
闫银松（香港理工大学）
P06
Adversarial Robust Generalization of Graph Neural Networks
曹畅（华中农业大学）
P07
An Efficient SE(p)-Invariant Transport Metric Driven by Polar Transport Discrepancy-based Representation
林俊一（中国人民大学）、薛敦耀（中国人民大学）
P08
AtC: Aggregate-then-Calibrate for Human-centered Assessment
李欣桐（中国人民大学）
P09
Attention-Only Transformers via Unrolled Subspace Clustering
汪鹏（澳门大学）
P10
Beyond Squared Error: Exploring Loss Design for Enhanced Training of Generative Flow Networks
胡睿（清华大学）
P11
Bias Pretends to Be Truth: How Spurious Correlations Undermine Hallucination Detection in LLMs
董亦骐（清华大学）
P12
Can MLLMs Understand the Deep Implication Behind Chinese Images?
张辰皓（华中科技大学，上海人工智能实验室）
P13
Continuous soft actor critic：An off-policy learning method robust to time discretization
韩慧敏（山东大学）
P14
CR-Net: Scaling Parameter-Efficient Training with Cross-Layer Low-Rank Structure
孔博傲（北京大学）
P15
DAGs as Minimal I-maps for the Induced Models of Causal Bayesian Networks under Conditioning
谢向东（安徽大学）
P16
Degradation-aware residual-conditioned optimal transport for unified image restoration
唐晓乐（西安交通大学）
P17
Diversity-Enhanced and Classification-Aware Prompt Learning for Few-Shot Learning via Stable Diffusion
常高钦（西安交通大学）
P18
Doubly Robust Alignment for Large Language Models
叶锴（LSE）
P19
DualOptim: Enhancing Efficacy and Stability in Machine Unlearning with Dual Optimizers
钟旭炀（香港城市大学）
P20
Functional Scaling Laws in Kernel Regression: Loss Dynamics and Learning Rate Schedules
李柄辉（北京大学）
P21
Functional Virtual Adversarial Training for Semi-Supervised Time Series Classification
潘庆一（清华大学）
P22
GraphMaster: Automated Graph Synthesis via LLM Agents in Data-Limited Environments
杜恩俊（香港科技大学（广州））
P23
High-Dimensional Tensor Regression with Oracle Properties
赵子平（上海科技大学）
P24
How does Labeling Error Impact Contrastive Learning? A Perspective from Data Dimensionality Reduction
陈君（华中农业大学）
P25
Improving Model Representation and Reducing KV Cache via Skip Connections with First Value Heads
吴洲同（北京大学）
P26
Jailbreak and Guard Aligned Language Models with Only Few In-Context Demonstrations
魏泽明（北京大学）
P27
Joint Velocity-Growth Flow Matching for Single-Cell Dynamics Modeling
王东熠（西安交通大学）
P28
Learning Continuous Wasserstein Barycenter Space for Generalized All-in-One Image Restoration
徐家怿（西安交通大学）
P29
Long-range Modeling and Processing of Multimodal Event Sequences
钟轶伦（中国人民大学）
P30
MIHC: Multi-View Interpretable Hypergraph Neural Networks with Information Bottleneck for Chip Congestion Prediction
张泽玥（中国人民大学）
P31
Mixture of Hidden-Dimensions: Not All Hidden-States' Dimensions are Needed in Transformer
陈奕龙（中国科学院）
P32
On the Surprising Effectiveness of a Single Global Merging in Decentralized Learning
朱同天（浙江大学）
P33
Preference-CFR: Beyond Nash Equilibrium for Better Game Strategies
居奇（华中科技大学）
P34
ProofAug: Efficient Neural Theorem Proving via Fine-grained Proof Structure Analysis
刘豪雄（蚂蚁集团）
P35
Reasoning Language Model Inference Serving Unveiled: An Empirical Study
伍俊潘（UW-Madison）
P36
Robust Trusted Conflictive Multiview Collaborative Contrastive Learning
张楠（温州大学）
P37
SE-Agent: Self-Evolution Trajectory Optimization in Multi-Step Reasoning with LLM-Based Agents
郭毅芙（华南师范大学）
P38
Sharpness-Aware Minimization Efficiently Selects Flatter Minima Late in Training
周展鹏（上海交通大学）
P39
Soccratic-Geo: Synthetic Data Generation and Geometric Reasoning via Multi-Agent Interaction
焦政博（上海财经大学）
P40
Stochastic Smoothed Primal-Dual Algorithms for Nonconvex Optimization with Linear Inequality Constraints
黄瑞川（中国科学技术大学）
P41
Suspicious Alignment of SGD: A Fine-Grained Step Size Condition Analysis
邓深洋（Dartmouth College）
P42
Task Generalization with Autoregressive Compositional Structure: Can Learning from D Tasks Generalize to D^T Tasks?
张华清（清华大学）
P43
Theoretical Modeling of LLM Self-Improvement Training Dynamics Through Solver-Verifier Gap
孙逸凡（上海财经大学）
P44
Time-uniform and Asymptotic Confidence Sequence of Quantile under Local Differential Privacy
蔡乐衡（清华大学）
P45
Trained Mamba Emulates Online Gradient Descent in In-Context Linear Regression
蒋佳锐（哈尔滨工业大学（深圳））
P46
uniINF: Best-of-both-worlds algorithm for parameter-free heavy-tailed MABs
陈禹（清华大学）
P47
Your Pre-trained LLM is Secretly an Unsupervised Confidence Calibrator
罗贝尔（南方科技大学）
P48
Understanding LLM Behaviors via Compression: Data Generation, Knowledge Acquisition and Scaling Laws
王少文（清华大学）
P49
Understanding the Forgetting of (Replay-based) Continual Learning via Feature Learning: Angle Matters
万弘毅（哈尔滨工业大学（深圳））
P50
MoFO: Momentum-Filtered Optimizer for Mitigating Forgetting in LLM Fine-Tuning
陈玉鹏（香港中文大学(深圳) / 深圳河套学院）、王森淼（香港中文大学（深圳））