1k citations · 2.2k across the 96 of their papers we have counts for
99 papers
Correcting or Continuing? What On-Policy Distillation Learns from Student Errors
Shuo Yang, Zheyu Zhang, Mahito Sugiyama +1
On-policy distillation (OPD) trains a student on its own rollouts under a frozen teacher, mitigating the distribution mismatch of off-policy supervision. While prior work has prima…
Relational Learning as World Modeling
Zheyu Zhang, Gjergji Kasneci
Relational learning typically predicts from a database state, but decision making asks a different question: how will that state change under an action? In operational systems, the…
High-Dimensional Online Change Point Detection with Adaptive Thresholding and Interpretability
Sven Jacob, Bardh Prenkaj, Weijia Shao +1
Change point detection (CPD) identifies abrupt and significant changes in sequential data, with applications in human activity recognition, financial markets, cybersecurity, manufa…
When Do Explanations Help In-Context Learning? A Comparative Study of Natural Language Explanation Types and Faithfulness
Mahdi Dhaini, Adam Dejl, Juraj Vladika +3
Natural language explanations (NLEs) are increasingly used as inputs, for example, as few-shot rationales that influence model behavior in in-context learning (ICL). However, it re…
Safety Cost of Steering Vectors Is Separable and Reducible
Yuxiao Li, Gjergji Kasneci
Steering vectors are a lightweight tool for controlling LLM behavior. However, emerging evidence shows that steering vectors can unintentionally compromise a model's safety mechani…
Beyond Binary Rewards: A Comparative Study of Reward Design for Reinforcement Unlearning
Efstratios Zaradoukas, Davide Gabrielli, Bardh Prenkaj +1
Machine unlearning seeks to selectively remove specific knowledge from trained language models without full retraining, a growing necessity under privacy regulations such as GDPR a…
The safety failures we are not instrumenting: a perspective on hidden safety-critical challenges in modern AI systems
Gjergji Kasneci, Enkelejda Kasneci
Current AI safety discourse still focuses disproportionately on visible failures, including obvious harms, dramatic misuse, and hypothetical catastrophic scenarios. That focus is i…
Every Eval Ever: A Unifying Schema and Community Repository for AI Evaluation Results
Jan Batzner, Sree Harsha Nelaturu, Damian Stachura +45
AI evaluations are widely used for testing and understanding progress. However, the diverse evaluators bring with them inconsistencies that challenge analysis and comparison. First…
Auditing LLM-Governed Social Robots with Culture-Specific Moral Gradients
Carmen Ng, Gjergji Kasneci
LLM-governed social robots increasingly decide who receives real-world assistance first. As prioritization norms vary across cultures by age, status, and group size, failure to cal…
Consolidating Rewarded Perturbations for LLM Post-Training
Zheyu Zhang, Shuo Yang, Gjergji Kasneci
Post-training of language models is commonly framed as a sample-score-update loop implemented by gradient descent. A recent line of work, exemplified by RandOpt, relocates this loo…
Sycophancy is an Educational Safety Risk: Why LLM Tutors Need Sycophancy Benchmarks
Enkelejda Kasneci, Gjergji Kasneci
This position paper argues that effective tutoring requires corrective friction: surfacing misconceptions and challenging them supportively to drive conceptual change. Yet preferen…
Active Tabular Augmentation via Policy-Guided Diffusion Inpainting
Zheyu Zhang, Shuo Yang, Bardh Prenkaj +1
Generative tabular augmentation is appealing in data-scarce domains, yet the prevailing focus on distributional fidelity does not reliably translate into better downstream models.…
Where Paths Split: Localized, Calibrated Control of Moral Reasoning in Large Language Models
Chenchen Yuan, Zheyu Zhang, Gjergji Kasneci
Large language models often display heterogeneous moral preferences across settings. We study inference-time steering toward a desired ethical framework while preserving general co…
SAGE: Sparse Adaptive Guidance for Dependency-Aware Tabular Data Generation
Shuo Yang, Zheyu Zhang, Bardh Prenkaj +1
Generating high-fidelity synthetic tabular data remains a critical challenge for enhancing data availability in privacy-sensitive and low-resource domains. Recent approaches levera…
TabSCM: A practical Framework for Generating Realistic Tabular Data
Sven Jacob, Bardh Prenkaj, Weijia Shao +1
Most tabular-data generators match marginal statistics yet ignore causal structure, leading downstream models to learn spurious or unfair patterns. We present TabSCM, a mixed-type…
Analysing the Safety Pitfalls of Steering Vectors
Yuxiao Li, Alina Fastowski, Efstratios Zaradoukas +2
Activation steering has emerged as a powerful tool to shape LLM behavior without the need for weight updates. While its inherent brittleness and unreliability are well-documented,…
Moral Lenses, Political Coordinates: Towards Ideological Positioning of Morally Conditioned LLMs
Chenchen Yuan, Bolei Ma, Zheyu Zhang +3
While recent research has systematically documented political orientation in large language models (LLMs), existing evaluations rely primarily on direct probing or demographic pers…
Reinforcement Unlearning via Group Relative Policy Optimization
Efstratios Zaradoukas, Bardh Prenkaj, Gjergji Kasneci
During pretraining, LLMs inadvertently memorize sensitive or copyrighted data, posing significant compliance challenges under legal frameworks like the GDPR and the EU AI Act. Fulf…
Sycophancy Claims about Language Models: The Missing Human-in-the-Loop
Jan Batzner, Volker Stocker, Stefan Schmid +1
Sycophantic response patterns in Large Language Models (LLMs) have been increasingly claimed in the literature. We review methodological challenges in measuring LLM sycophancy and…
Whose Personae? Synthetic Persona Experiments in LLM Research and Pathways to Transparency
Jan Batzner, Volker Stocker, Bingjun Tang +4
Synthetic personae experiments have become a prominent method in Large Language Model alignment research, yet the representativeness and ecological validity of these personae vary…
Injecting Falsehoods: Adversarial Man-in-the-Middle Attacks Undermining Factual Recall in LLMs
Alina Fastowski, Bardh Prenkaj, Yuxiao Li +1
LLMs are now an integral part of information retrieval. As such, their role as question answering chatbots raises significant concerns due to their shown vulnerability to adversari…
Assessing the Real-World Utility of Explainable AI for Arousal Diagnostics: An Application-Grounded User Study
Stefan Kraft, Andreas Theissler, Vera Wienhausen-Wilke +2
Artificial intelligence (AI) systems increasingly match or surpass human experts in biomedical signal interpretation. However, their effective integration into clinical practice re…
Structured Universal Adversarial Attacks on Object Detection for Video Sequences
Sven Jacob, Weijia Shao, Gjergji Kasneci
Video-based object detection plays a vital role in safety-critical applications. While deep learning-based object detectors have achieved impressive performance, they remain vulner…
CURE: Controlled Unlearning for Robust Embeddings -- Mitigating Conceptual Shortcuts in Pre-Trained Language Models
Aysenur Kocak, Shuo Yang, Bardh Prenkaj +1
Pre-trained language models have achieved remarkable success across diverse applications but remain susceptible to spurious, concept-driven correlations that impair robustness and…