Stephan Günnemann

Sampling-aware Adversarial Attacks Against Large Language Models

Discrete Bayesian Sample Inference for Graph Generation

Reliability Scaling Laws for Quantized Large Language Models

Model Collapse Is Not a Bug but a Feature in Machine Unlearning for LLMs

Edit-Based Flow Matching for Temporal Point Processes

OmegAMP: Targeted AMP Discovery via Biologically Informed Generation

SAFT: Structure-Aware Fine-Tuning of LLMs for AMR-to-Text Generation

Exact Certification of Neural Networks and Partition Aggregation Ensembles against Label Poisoning

Adversarial Robustness of Continuous Time Dynamic Graphs

Generative Modeling with Bayesian Sample Inference

Anomaly-Gym: A Benchmark for Anomaly Detection in Embodied Agent Environments

Effective Data Pruning through Score Extrapolation

Byte Pair Encoding for Efficient Time Series Forecasting

Certifying Graph Neural Networks Against Label and Structure Poisoning

Long-Range Graph Wavelet Networks

Energy-Weighted Flow Matching: Unlocking Continuous Normalizing Flows for Efficient and Scalable Boltzmann Sampling

A Probabilistic Perspective on Unlearning and Alignment for Large Language Models

Privacy Amplification by Structured Subsampling for Deep Differentially Private Time Series Forecasting

Learning Equivariant Non-Local Electron Density Functionals

Exact Certification of (Graph) Neural Networks Against Label Poisoning

Provably Reliable Conformal Prediction Sets in the Presence of Data Poisoning

What Expressivity Theory Misses: Message Passing Complexity for GNNs

Modeling Microenvironment Trajectories on Spatial Transcriptomics with NicheFlow

MAGNet: Motif-Agnostic Generation of Molecules from Scaffolds

Graph Neural Networks for Edge Signals: Orientation Equivariance and Invariance

Unlocking Point Processes through Point Set Diffusion

Consistent Sampling and Simulation: Molecular Dynamics with Energy-Based Diffusion Models

Flow Matching with Gaussian Process Priors for Probabilistic Time Series Forecasting

The Geometry of Refusal in Large Language Models: Concept Cones and Representational Independence

REINFORCE Adversarial Attacks on Large Language Models: An Adaptive, Distributional, and Semantic Objective

Provable Robustness of (Graph) Neural Networks Against Data Poisoning and Backdoors

Lift Your Molecules: Molecular Graph Generation in Latent Euclidean Space

Joint Relational Database Generation via Graph-Conditional Diffusion Models

TreeGen: A Bayesian Generative Model for Hierarchies

Uncertainty Estimation for Heterophilic Graphs Through the Lens of Information Theory

Enforcing Latent Euclidean Geometry in Single-Cell VAEs for Manifold Interpolation

Enforcing Latent Euclidean Geometry in VAEs for Statistical Manifold Interpolation

Adversarial Robustness of Graph Transformers

Efficient Time Series Processing for Transformers and State-Space Models through Token Merging

UnHiPPO: Uncertainty-aware Initialization for State Space Models

Efficient Time Series Processing for Transformers and State-Space Models through Token Merging

A Unified Approach Towards Active Learning and Out-of-Distribution Detection