3 papers
cs.LG2026
MechRL: Reinforcement Learning Agents Perform Circuit Discovery for Mechanistic Interpretability
Barsat Khadka
Mechanistic interpretability seeks to explain a model's behaviour by finding its circuit: the sparse subgraph of the model's computation that is causally responsible for it. Automa…
cs.LG2026
SwiftCTS: Fast Cross-Design Prediction and Pareto Optimization of Clock Tree Metrics via Few-Shot Calibration
Barsat Khadka, Kawsher Roxy, Md Rubel Ahmed
Clock Tree Synthesis (CTS) is a computationally expensive stage in the physical design flow, requiring iterative EDA tool invocations to navigate a vast configuration space for opt…
cs.CR2026
Filter-then-Verify: A Multiphase GNN and ModernBERT Framework for Social Engineering Detection in Email Networks
Barsat Khadka, Prasant Koirala, Kshitiz Neupane +1
Social engineering attacks exploit human trust rather than software vulnerabilities, making them difficult to detect using conventional filters. We propose a two-stage filter-then-…