collaborators

7 papers

cs.CL2026

Detection Without Correction: A Robust Asymmetry in Activation-Based Hallucination Probing

Dip Roy, Rajiv Misra, Sanjay Kumar Singh +1

Activation-based linear probing is widely proposed as a method for both detecting and correcting hallucinations in autoregressive language models. We present an empirical study acr…

cs.LG2026

Posterior-Calibrated Causal Circuits in Variational Autoencoders: Why Image-Domain Interpretability Fails on Tabular Data

Dip Roy, Rajiv Misra, Sanjay Kumar Singh +1

Although mechanism-based interpretability has generated an abundance of insight for discriminative network analysis, generative models are less understood -- particularly outside o…

cs.LG2026

A Multi-Level Causal Intervention Framework for Mechanistic Interpretability in Variational Autoencoders

Dip Roy, Rajiv Misra, Sanjay Kumar Singh +1

Understanding how generative models represent and transform data is a foundational problem in deep learning interpretability. While mechanistic interpretability of discriminative a…

cs.LG2026

MemGuard-Alpha: Detecting and Filtering Memorization-Contaminated Signals in LLM-Based Financial Forecasting via Membership Inference and Cross-Model Disagreement

Anisha Roy, Dip Roy

Large language models (LLMs) are increasingly used to generate financial alpha signals, yet growing evidence shows that LLMs memorize historical financial data from their training…

cs.LG2026

Fundamental Limits of Neural Network Sparsification: Evidence from Catastrophic Interpretability Collapse

Dip Roy, Rajiv Misra, Sanjay Kumar Singh

Extreme neural network sparsification (90% activation reduction) presents a critical challenge for mechanistic interpretability: understanding whether interpretable features surviv…

cs.CV2026

Mechanistic Interpretability of Diffusion Models: Circuit-Level Analysis and Causal Validation

Dip Roy

We present a quantitative circuit-level analysis of diffusion models, establishing computational pathways and mechanistic principles underlying image generation processes. Through…