Research & Academia

Published work on AI safety, adversarial robustness, and machine learning systems.

Papers

Perturbation-Based Generation Profiling Detects Covert AI Agent Attacks Where Token-Level Statistics Fail

Yatharth Maheshwari, Arka Dash, Abhineet Som

Unsupervised attack detection via cross-perturbation generation profiling — 0.94–1.00 AUROC across six LLMs (8B–32B) with zero training data, where token-level baselines plateau at 0.56–0.83.

Education

IIT Madras

Bachelor of Science in Data Science and Programming · 2023 – 2027

CGPA: 8.8/10 · Remote (Dual Degree)

Kalyani Government Engineering College

Bachelor of Technology in Computer Science · 2023 – 2027

CGPA: 8.4/10 · Kalyani, India

Recent Writing

Nothing here yet — check back later.