Manit Baser

Disentangling Language Effects on Tool Admission and Post-Retrieval Safety

A Nawal, M Baser, M Gurusamy

Preprint

Before Retrieval Begins: Safety and Tool-Use Decisions in LLM Agents

A Nawal, M Baser, M Gurusamy

Preprint

The Fragility of Trigger-Tag Mechanisms for Misuse Detection in Open-Weight LLMs

T Aremu_*, M Baser_*, M Gurusamy, N Lukas, DM Divakaran

Preprint

The Tokens Remember: When Tokenization Bypasses Knowledge Editing and Unlearning

M Baser, A Nawal, DM Divakaran, M Gurusamy

Preprint

Relevance as a Vulnerability: How Web Retrieval Degrades Safety Alignment in LLM Agents

A Nawal, M Baser, M Gurusamy

EMNLP-Findings 2026

CLaRE-ty Amid Chaos: Quantifying Representational Entanglement to Predict Ripple Effects in LLM Editing

M Baser, A Yildiz, DM Divakaran, M Gurusamy

ACL-Findings 2026

ThinkEval: Practical Evaluation of Knowledge Leakage in LLM Editing using Thought-based Knowledge Graphs

M Baser, DM Divakaran, M Gurusamy

TMLR 2026

DarKnight: Retrieval-Guided Agentic Framework for Breaking Dark Web CAPTCHA Gates

B Wu, ZY Yeo, D Tan, M Baser, L Csikor, DM Divakaran

APWG eCrime 2026

Step-by-Step Reasoning Attack: Revealing 'Erased' Knowledge in Large Language Models

Y Sinha, M Baser, M Mandal, DM Divakaran, M Kankanhalli

Preprint

Three-Tier Indirect Tracing Model for Enhancing Epidemic Surveillance

G Bansal, M Baser, V Chamola

IEEE IoT Magazine 2024