1.3k citations · 3.2k across the 145 of their papers we have counts for
173 papers
EAIR: Interaction Risks in Enterprise LLM Multi-Agent Systems
Yuchen Ma, Swadheen Shukla, Paulo Salem +5
LLM agents increasingly divide enterprise work across roles with different goals, authority, access, and information. Because each agent often sees only a local snapshot of the ent…
OrthoGen: A Generative Orthogonal Learner for Time-Varying Treatments
Tomàs Garriga, Valentyn Melnychuk, Konstantin Hess +4
Estimating conditional distributional potential outcomes (CDPOs) over time is important in medicine (e.g., to estimate patient-specific risks under different treatment sequences).…
Superconducting Diode Effect due to the Inverse Spin Hall Effect in Josephson Junctions with Extrinsic Spin-Orbit Interaction
Yositake Takane, Aurélien Manchon, Gen Tatara
The spin density and Josephson current at equilibrium are studied numerically in a superconductor-normal metal-superconductor (SNS) junction with extrinsic spin-orbit interaction i…
Prediction-powered Neural Architecture Search
Pascal Janetzky, Yuxin Wang, Michael Klar +1
Evaluating candidate architectures in neural architecture search (NAS) faces an inherent trade-off: on the one hand, reliable performance labels are limited because training and ev…
Representation Learning for Exact Preimages
Konstantin Hess, Stefan Feuerriegel
Modern neural predictors can model highly nonlinear maps, but many scientific and engineering tasks require reasoning in the opposite direction: given a performance or safety level…
Context-dependent agent evaluation with orthogonal equilibrium learning
Haorui Ma, Zehua Zang, Jiangmeng Li +3
Many applications require to evaluate agents under contextual information (e.g., a prompt, task, or user group). We study how to perform such context-dependent agent evaluation fro…
Path-specific harm decomposition: A partial identification framework
Ruizi Yan, Dennis Frauen, Maresa Schröder +1
A central goal when designing treatment policies is often to "do no harm", that is, to avoid interventions that improve average outcomes while worsening outcomes for some individua…
Reward Hacking Challenges Oversight of Autonomous Research Agents
Yue Huang, Zhangchen Xu, Yuchen Ma +12
Autonomous research agents can design experiments, evaluate results, and write reports, giving them control over both a scientific result and the evidence used to support it. This…
Designing Against Deskilling: Metacognitive Feedback Reduces Cognitive Offloading to LLM Assistants
Sebastian Maier, Kai Schwabe, Manuel Schneider +1
Cognitive offloading to AI can reduce opportunities to practice skills, creating risks of deskilling. However, it remains unclear how to prevent deskilling without restricting acce…
XAI-Arena: Can LLMs Assess the Quality of XAI Explanations?
Yanfei Hu Fleischhauer, Alona Zharova, Nadja Klein +1
Evaluating the quality of explanations produced by explainable AI (XAI) methods remains challenging because existing approaches often rely on subjective human judgment, limiting re…
Beyond Surface Cues: Disentangling Sociocultural Signals in Multilingual LLMs
Yuanjun Feng, Tanzhou Liu, Stefan Feuerriegel +1
Multilingual LLM outputs can vary across sociocultural contexts. However, evidence of cultural grounding can be misleading: identity labels may be inferred from explicit or indirec…
LLM-SoccerArena: Benchmarking LLMs on Real-World Predictions in Sports
Jonas Schröder, Jonas Schweisthal, Oliver Müller +2
Large language models (LLMs) increasingly support decisions about uncertain future events, yet evaluating their ability to forecast real-world outcomes remains difficult. In partic…
AutoSynthesis: An agentic system for automated meta-analysis
Moein Taherinezhad, Sebastian Maier, Gerardo Vitagliano +2
Evidence synthesis is crucial for turning primary research into reliable knowledge for science, medicine, education, and policy. Yet, quantitative evidence synthesis remains largel…
Hyperlocal monitoring of urban activity reveals responses to heat exposure
Dominique Geissler, Felix Creutzig, Ramit Debnath +1
Rising temperatures create new challenges for local heat adaptation. Yet, it remains unclear how urban activity changes during hot periods and which urban environments people conce…
Automated reproducibility assessments in the social and behavioral sciences using large language models
Tobias Holtdirk, Pietro Marcolongo, Anna Steinberg Schulten +7
Reproducibility in the social and behavioral sciences is typically evaluated by independent researchers who reanalyze the original data to assess whether the published findings can…
OncoSynth: Synthetic data generation for treatment effect estimation in oncology
Octavia-Andreea Ciora, Julian Welzel, Dennis Frauen +6
In oncology, access to patient-level data is often restricted. Synthetic data provides an alternative for analyzing treatment effectiveness, but existing methods for synthetic data…
Gender-based discrepancies in the algorithmic delivery of political ads on social media
Dominik Bär, Francesco Corso, Gianmarco De Francisci Morales +2
Social media has become a key channel for political advertising during election campaigns. However, algorithmic biases in the delivery of these ads may distort the public's exposur…
Causal methods for LLM development and evaluation
Dennis Frauen, Marie Brockschmidt, Konstantin Hess +10
Large language model (LLM) development is currently driven by large-scale empirical iteration over data mixtures, reward models, routing strategies, and evaluation pipelines. Here,…
Adaptive Experimentation for Censored Survival Outcomes
Yuxin Wang, Dennis Frauen, Jonas Schweisthal +3
Adaptive experimentation enables efficient estimation of causal effects, but existing methods are not designed for survival data with censoring, where event times are only partiall…
Continual Learning of Domain-Invariant Representations
Pascal Janetzky, Tobias Schlagenhauf, Stefan Feuerriegel
Continual learning (CL) aims to train models sequentially over multiple domains without forgetting previously learned knowledge. However, existing CL methods optimize for in-domain…
Amortizing Causal Sensitivity Analysis via Prior Data-Fitted Networks
Emil Javurek, Dennis Frauen, Marie Brockschmidt +2
Causal sensitivity analysis aims to provide bounds for causal effect estimates in the presence of unobserved confounding. However, existing methods for causal sensitivity analysis…
ConfoundingSHAP: Quantifying confounding strength in causal inference
Marie Brockschmidt, Santo M. A. R. Thies, Maresa Schröder +5
In causal inference, confounders are variables that influence both treatment decisions and outcomes. However, unlike as in randomized clinical trials, the treatment assignment mech…
SkillGen: Verified Inference-Time Agent Skill Synthesis
Yuchen Ma, Yue Huang, Han Bao +5
Skills are a promising way to improve LLM agent capabilities without retraining, while keeping the added procedure reusable and controllable. However, high-quality skills are still…
NARRA-Gym for Evaluating Interactive Narrative Agents
Yue Huang, Yuchen Ma, Jiayi Ye +14
Interactive narrative tasks require LLMs to sustain a coherent, evolving story while adapting to a user over multiple turns. However, suitable benchmarks for this setting are limit…