activity
20242026
collaborators

5 papers

cs.LG2026

Detecting Contaminated Code-Generation Prompt Batches via Influence Functions

Francesco Quinzan, Noor Munir, Yishun Lu +1

Large language models (LLMs) are increasingly used for code generation, yet they remain vulnerable to prompts that elicit insecure implementations. Existing defenses typically rely…

cs.LG2025

Training Instabilities Induce Flatness Bias in Gradient Descent

Lawrence Wang, Stephen J. Roberts

Classical analyses of gradient descent (GD) define a stability threshold based on the largest eigenvalue of the loss Hessian, often termed sharpness. When the learning rate lies be…

cs.LG2025

Pruning Cannot Hurt Robustness: Certified Trade-offs in Reinforcement Learning

James Pedley, Benjamin Etheridge, Stephen J. Roberts +1

Reinforcement learning (RL) policies deployed in real-world environments must remain reliable under adversarial perturbations. At the same time, modern deep RL agents are heavily o…

cs.LG2025

Beyond Accuracy: EcoL2 Metric for Sustainable Neural PDE Solvers

Taniya Kapoor, Abhishek Chandra, Anastasios Stamou +1

Real-world systems, from aerospace to railway engineering, are modeled with partial differential equations (PDEs) describing the physics of the system. Estimating robust solutions…

cs.LG2024

Can Stability be Detrimental? Better Generalization through Gradient Descent Instabilities

Lawrence Wang, Stephen J. Roberts

Traditional analyses of gradient descent optimization show that, when the largest eigenvalue of the loss Hessian - often referred to as the sharpness - is below a critical learning…