5 papers
Measuring and Mitigating Solution Mode Collapse in RLVR
Liv G. d'Aliberti, Marwa Abdulhai, Sofiia Druchyna +2
A language model (LM) can usually answer the same question in more than one way, but reinforcement learning with verifiable rewards (RLVR) is indifferent to which correct answer a…
Behavior-Consistent Deep Reinforcement Learning
Marcel Hussing, Liv G. d'Aliberti, Claas Voelcker +2
Reinforcement learning (RL) often exhibits high variance across training runs, leading to unreliable performance and posing a major challenge to deployment in real-world domains. I…
The Illusion of Insight in Reasoning Models
Liv G. d'Aliberti, Manoel Horta Ribeiro
Do reasoning models have "Aha!" moments? Prior work suggests that models like DeepSeek-R1-Zero undergo sudden mid-trace realizations that lead to accurate outputs, implying an intr…
BlindFL: Segmented Federated Learning with Fully Homomorphic Encryption
Evan Gronberg, Liv d'Aliberti, Magnus Saebo +1
Federated learning (FL) is a popular privacy-preserving edge-to-cloud technique used for training and deploying artificial intelligence (AI) models on edge devices. FL aims to secu…
Privacy-Enhancing Technologies for Artificial Intelligence-Enabled Systems
Liv d'Aliberti, Evan Gronberg, Joseph Kovba
Artificial intelligence (AI) models introduce privacy vulnerabilities to systems. These vulnerabilities may impact model owners or system users; they exist during model development…