Neelu Verma

Risk-Calibrated Prompting: Selective Answering for Robustness Under Prompt Perturbations

Large language models (LLMs) can produce fluent responses that remain unreliable when a question is rephrased, surrounded by irrelevant context, or intrinsically ambiguous. Most prompt-based hallucination mitigation methods optimize answer quality or average factual error, whereas deployment also requires a decision ab...

Retrieval-augmented chart-grounded fact verification with self-consistent vision–language reasoning