1 paper
Muhammad Ashad Kabir, Kawsar Ahmed, Md. Osama
Large language models (LLMs) have shown strong performance in text generation tasks, yet their effectiveness on headline generation remains sensitive to how input context is select…