5 papers
Not All Claims Are Equally Risky: FACTOR for Adaptive Verification in Factual Long-Form Generation
Areeba Hassan, Arooj Kausar, Syeda Kisaa Fatima +2
Large Language Models (LLMs) generate fluent long-form text, however, often add unsupported factual claims. Existing verification techniques improve factuality by grounding generat…
PRIME: Evaluating Prompt Resolution Under Incompatible Instructions in LLMs
Tehreem Javed, Shumaim Fatimah, Masooma Bakhtiari +2
Large language models (LLMs) often encounter conflicting prompts, although current instruction following benchmarks assess those meta-instructions in isolation, limiting the insigh…
Subjective Question Generation and Answer Evaluation using NLP
G. M. Refatul Islam, Safwan Shaheer, Yaseen Nur +1
Natural Language Processing (NLP) is one of the most revolutionary technologies today. It uses artificial intelligence to understand human text and spoken words. It is used for tex…
Beyond the Benchmark: Innovative Defenses Against Prompt Injection Attacks
Safwan Shaheer, G. M. Refatul Islam, Mohammad Rafid Hamid +1
In this fast-evolving area of LLMs, our paper discusses the significant security risk presented by prompt injection attacks. It focuses on small open-sourced models, specifically t…
Detecting Prompt Injection Attacks Against Application Using Classifiers
Safwan Shaheer, G. M. Refatul Islam, Mohammad Rafid Hamid +3
Prompt injection attacks can compromise the security and stability of critical systems, from infrastructure to large web applications. This work curates and augments a prompt injec…