Publications

AI safety, reasoning evaluation, multilingual NLP.

2026

Conversational Hallucination Drift: An Episodic Retrieval Framework and Error Taxonomy for Long-Term Memory Evaluation

Kamesh R, Sivadhanushya P · ICML 2026, Workshop on Failure Modes of Agentic AI

Global PIQA: Evaluating Physical Commonsense Reasoning Across 100+ Languages and Cultures

Kamesh R · NeurIPS 2026, Under review

2025

Probing Reasoning Flaws and Safety Hierarchies with Chain-of-Thought Difference Amplification

Kamesh R · NeurIPS 2025, LLM-Evals Workshop

Indian Grammatical Tradition-Inspired Universal Semantic Representation Bank (USR Bank 1.0)

Kamesh R et al. · AACL-IJCNLP 2025, BHASHA Workshop