projects
-
Investigating Model Organism Robustness to Fine-Tuning
Construction method impact on quirk persistence during untargeted supervised fine-tuning.
-
Refusal Direction Persistence Through Supervised Fine-Tuning
Supervised fine-tuning changes refusal behaviour while preserving a causally effective refusal direction.
-
Evaluating Refusal and Jailbreak Behaviour in Gemma 3
Behaviour shifts of Gemma 3 27B under safety-focused and role-play instructions.
-
Tapering Away Normalization in Transformers
Gated removal of normalization in Transformers for stable training and faster inference.
-
Detecting RAG Hallucinations with Output Probabilities and Attention
A short study of whether token probabilities and attention can flag unsupported RAG answers.
-
Parameter-efficient Adaptation of Tokenizer-free Byte Latent Transformer
Adapting a tokenizer-free architecture to new languages by retraining only the ~4% "interface" modules and keeping the core transformer fixed.
-
Circuit Tracing Walk-or-Drive Decisions
Provide a single word answer. Should I walk/drive to my destination which is {distance} m away
-
Safe Explicit MPC by Training Neural Networks Through Constrained Optimization
Fast explicit MPC policies learned while enforcing closed-loop safety during neural network training.
-
Analysis and Local Convergence Proof of a Constrained Optimization Algorithm for Training Neural Networks
Local convergence analysis of a primal-dual Adam method for constrained neural network training.
-
Notes on Interpretability
An evolving taxonomy of interpretability research
-
Thermal Runaway Avoidance Using Hamilton–Jacobi Reachability and Model Predictive Control
Combining Hamilton–Jacobi reachability with MPC to avoid thermal runaway in exothermic batch processes.
-
WNNM Image Denoising
Python implementation of weighted nuclear norm minimization (WNNM) and comparison against Gaussian, bilateral, and NLM baselines.
-
Invariance Constraints for Computational Lithography
Injecting expert invariances into deep models (CNNs) with exact-fit constraints on critical samples.
-
Inverse Modeling for Metrology
Faster, more accurate inverse modeling for semiconductor metrology.