3 citations · 7 across the 12 of their papers we have counts for
5 papers · 1 filter
SimHack: Predicting the onset of detectable reward hacking before RL training
Junwei Deng, Alex Bie, Natalia Ponomareva
When training LLMs to code with reinforcement learning, reward hacking is often detectable but difficult to prevent, since (a) environment hardening is increasingly difficult as tr…
Two For The Price Of One: Curvature-Aware Pruning Preserves Influence Functions Too
Mehdi Makni, Natalia Ponomareva, Rahul Mazumder
Post-training pruning enables efficient deployment, but conventional methods do not explicitly preserve internal-gradient alignment between sparse and dense models. We study novel…
ContinuousBench: Can Differentially Private Synthetic Text Improve Capabilities?
Peihan Liu, Lucas Rosenblatt, Weiwei Kong +7
Differentially private (DP) text synthesis promises to unlock sensitive corpora for model training, but it remains unclear whether DP synthetic data transmits genuinely new knowled…
Decomposing Private Image Generation via Coarse-to-Fine Wavelet Modeling
Jasmine Bayrooti, Weiwei Kong, Natalia Ponomareva +3
Generative models trained on sensitive image datasets risk memorizing and reproducing individual training examples, making strong privacy guarantees essential. While differential p…
Privately Fine-Tuned LLMs Preserve Temporal Dynamics in Tabular Data
Lucas Rosenblatt, Peihan Liu, Ryan McKenna +1
Research on differentially private synthetic tabular data has largely focused on independent and identically distributed rows where each record corresponds to a unique individual.…