Copertina

Diffusion LLMs are Natural Adversaries for any LLM

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Lüdke, David, Wollschläger, Tom, Ungermann, Paul, Günnemann, Stephan, Schwinn, Leo
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Closing the Distribution Gap in Adversarial Training for LLMs
di: Hu, Chengzhi, et al.
Pubblicazione: (2026)

Joint Relational Database Generation via Graph-Conditional Diffusion Models
di: Ketata, Mohamed Amine, et al.
Pubblicazione: (2025)

Adversarial Alignment for LLMs Requires Simpler, Reproducible, and More Measurable Objectives
di: Schwinn, Leo, et al.
Pubblicazione: (2025)

Flow Matching with Gaussian Process Priors for Probabilistic Time Series Forecasting
di: Kollovieh, Marcel, et al.
Pubblicazione: (2024)

Efficient Adversarial Training in LLMs with Continuous Attacks
di: Xhonneux, Sophie, et al.
Pubblicazione: (2024)

Sampling-aware Adversarial Attacks Against Large Language Models
di: Beyer, Tim, et al.
Pubblicazione: (2025)

Adversarial Robustness of Graph Transformers
di: Foth, Philipp, et al.
Pubblicazione: (2024)

Model Collapse Is Not a Bug but a Feature in Machine Unlearning for LLMs
di: Scholten, Yan, et al.
Pubblicazione: (2025)

A Probabilistic Perspective on Unlearning and Alignment for Large Language Models
di: Scholten, Yan, et al.
Pubblicazione: (2024)

Extracting Unlearned Information from LLMs with Activation Steering
di: Seyitoğlu, Atakan, et al.
Pubblicazione: (2024)

Soft Prompt Threats: Attacking Safety Alignment and Unlearning in Open-Source LLMs through the Embedding Space
di: Schwinn, Leo, et al.
Pubblicazione: (2024)

Byte Pair Encoding for Efficient Time Series Forecasting
di: Götz, Leon, et al.
Pubblicazione: (2025)

Efficient Time Series Processing for Transformers and State-Space Models through Token Merging
di: Götz, Leon, et al.
Pubblicazione: (2024)

What Expressivity Theory Misses: Message Passing Complexity for GNNs
di: Kemper, Niklas, et al.
Pubblicazione: (2025)

Energy-based Epistemic Uncertainty for Graph Neural Networks
di: Fuchsgruber, Dominik, et al.
Pubblicazione: (2024)

The Illusion of Certainty: Uncertainty Quantification for LLMs Fails under Ambiguity
di: Tomov, Tim, et al.
Pubblicazione: (2025)

Unlocking Point Processes through Point Set Diffusion
di: Lüdke, David, et al.
Pubblicazione: (2024)

Add and Thin: Diffusion for Temporal Point Processes
di: Lüdke, David, et al.
Pubblicazione: (2023)

Joint Out-of-Distribution Filtering and Data Discovery Active Learning
di: Schmidt, Sebastian, et al.
Pubblicazione: (2025)

Expressivity of Graph Neural Networks Through the Lens of Adversarial Robustness
di: Campi, Francesco, et al.
Pubblicazione: (2023)

Edit-Based Flow Matching for Temporal Point Processes
di: Lüdke, David, et al.
Pubblicazione: (2025)

Assessing Robustness via Score-Based Adversarial Image Generation
di: Kollovieh, Marcel, et al.
Pubblicazione: (2023)

Uncertainty Estimation for Heterophilic Graphs Through the Lens of Information Theory
di: Fuchsgruber, Dominik, et al.
Pubblicazione: (2025)

Effective Data Pruning through Score Extrapolation
di: Schmidt, Sebastian, et al.
Pubblicazione: (2025)

Localized Randomized Smoothing for Collective Robustness Certification
di: Schuchardt, Jan, et al.
Pubblicazione: (2022)

From Zero to Turbulence: Generative Modeling for 3D Flow Simulation
di: Lienen, Marten, et al.
Pubblicazione: (2023)

Uncertainty for Active Learning on Graphs
di: Fuchsgruber, Dominik, et al.
Pubblicazione: (2024)

Expressivity and Generalization: Fragment-Biases for Molecular GNNs
di: Wollschläger, Tom, et al.
Pubblicazione: (2024)

REINFORCE Adversarial Attacks on Large Language Models: An Adaptive, Distributional, and Semantic Objective
di: Geisler, Simon, et al.
Pubblicazione: (2025)

Unexplored flaws in multiple-choice VQA evaluations
di: Rosenthal, Fabio, et al.
Pubblicazione: (2025)

Lift Your Molecules: Molecular Graph Generation in Latent Euclidean Space
di: Ketata, Mohamed Amine, et al.
Pubblicazione: (2024)

Energy-Weighted Flow Matching: Unlocking Continuous Normalizing Flows for Efficient and Scalable Boltzmann Sampling
di: Dern, Niclas, et al.
Pubblicazione: (2025)

Attacking Large Language Models with Projected Gradient Descent
di: Geisler, Simon, et al.
Pubblicazione: (2024)

Certifiably Robust Encoding Schemes
di: Saxena, Aman, et al.
Pubblicazione: (2024)

Discrete Randomized Smoothing Meets Quantum Computing
di: Wollschläger, Tom, et al.
Pubblicazione: (2024)

Adversarial Attacks on Graph Neural Networks via Meta Learning
di: Zügner, Daniel, et al.
Pubblicazione: (2019)

A Scalable Multi-Task Model for Virtual Sensors
di: Götz, Leon, et al.
Pubblicazione: (2026)

Large-Scale Dataset Pruning in Adversarial Training through Data Importance Extrapolation
di: Nieth, Björn, et al.
Pubblicazione: (2024)

Interpolating Discrete Diffusion Models with Controllable Resampling
di: Kollovieh, Marcel, et al.
Pubblicazione: (2026)

Task-Awareness Improves LLM Generations and Uncertainty
di: Tomov, Tim, et al.
Pubblicazione: (2026)