AIConfPaper
← Search

Niklas Pfister

8 accepted papers

2026

Breaking Agent Backbones: Evaluating the Security of Backbone LLMs in AI Agents

Julia Bazinska, Max Mathys, Francesco Casucci, Mateo Rojas-Carulla, Xander Davies, Alexandra Souly +1
ICLR 2026poster

AI agents powered by large language models (LLMs) are being deployed at scale, yet we lack a systematic understanding of how the choice of backbone LLM affects agent security. The non-deterministic sequential nature of AI agents complicates security modeling, while the integration of traditional sof…

Cited by 0Source ↗Scholar ↗
2026

Many Experiments, Few Repetitions, Unpaired Data, and Sparse Effects: Is Causal Inference Possible?

Felix Schur, Niklas Pfister, Peng Ding, Sach Mukherjee, Jonas Peters
ICML 2026spotlight

In many applications, practical constraints prevent measuring covariates and outcomes on the same units, resulting in unpaired data. We study the problem of estimating causal effects under hidden confounding in the following unpaired data setting: we observe some covariates $X$ and an outcome $Y$ un…

Cited by 0Source ↗Scholar ↗
2025

Fast Estimation of Partial Dependence Functions using Trees

Jinyang Liu, Tessa Steensgaard, Marvin N. Wright, Niklas Pfister, Munir Hiabu
ICML 2025poster

Many existing interpretation methods are based on Partial Dependence (PD) functions that, for a pre-trained machine learning model, capture how a subset of the features affects the predictions by averaging over the remaining features. Notable methods include Shapley additive explanations (SHAP) whic…

Cited by 0Source ↗PDF ↗Scholar ↗
2025

Gandalf the Red: Adaptive Security for LLMs

Niklas Pfister, Václav Volhejn, Manuel Knott, Santiago Arias, Julia Bazinska, Mykhailo Bichurin +20
ICML 2025poster

Current evaluations of defenses against prompt attacks in large language model (LLM) applications often overlook two critical factors: the dynamic nature of adversarial behavior and the usability penalties imposed on legitimate users by restrictive defenses. We propose D-SEC (Dynamic Security Utilit…

Cited by 0Source ↗PDF ↗code ↗Scholar ↗
2025

Sparse Causal Effect Estimation using Two-Sample Summary Statistics in the Presence of Unmeasured Confounding

Shimeng Huang, Niklas Pfister, Jack Bowden
AISTATS 2025poster

Observational genome-wide association studies are now widely used for causal inference in genetic epidemiology. To maintain privacy, such data is often only publicly available as summary statistics, and often studies for the endogenous covariates and the outcome are available separately. This has ne…

Cited by 0Source ↗code ↗Scholar ↗
2024

Identifying Representations for Intervention Extrapolation

Sorawit Saengkyongam, Elan Rosenfeld, Pradeep Kumar Ravikumar, Niklas Pfister, Jonas Peters
ICLR 2024poster

The premise of identifiable and causal representation learning is to improve the current representation learning paradigm in terms of generalizability or robustness. Despite recent progress in questions of identifiability, more theoretical results demonstrating concrete advantages of these methods f…

Cited by 18Source ↗PDF ↗Scholar ↗
2022

Exploiting Independent Instruments: Identification and Distribution Generalization

Sorawit Saengkyongam, Leonard Henckel, Niklas Pfister, Jonas Peters
ICML 2022spotlight

Instrumental variable models allow us to identify a causal function between covariates $X$ and a response $Y$, even in the presence of unobserved confounding. Most of the existing estimators assume that the error term in the response $Y$ and the hidden confounders are uncorrelated with the instrumen…

Cited by 21Source ↗PDF ↗code ↗Scholar ↗