Papers by Shagufta Mehnaz
From Insight to Exploit: Leveraging LLM Collaboration for Adaptive Adversarial Text Generation (2025.findings-emnlp)
Copied to clipboard
| Challenge: | LLMs can provide substantial zero-shot performance on diverse tasks, but it is crucial to assess their robustness against adversarial inputs. |
| Approach: | They introduce Static Deceptor and Dynamic Deceptr to generate adversarial examples . they produce subtle and natural-looking adversarials that preserve semantic similarity to text . |
| Outcome: | The proposed attacks are based on two LLM-based attacks that generate natural-looking examples that deceive the target LLM. |
LogicEval: A Systematic Framework for Evaluating Automated Repair Techniques for Logical Vulnerabilities in Real-World Software (2026.acl-long)
Copied to clipboard
Syed Md Mukit Rashid, Abdullah Al Ishtiaq, Kai Tu, Yilu Dong, Tianwei Wu, Ali Ranjbar, Tianchang Yang, Najrin Sultana, Shagufta Mehnaz, Syed Rafiul Hussain
| Challenge: | Existing automated program-repair techniques focus on repairing memory corruptions, but they struggle with logical vulnerabilities because of their limited semantic understanding of the code and its expected behavior. |
| Approach: | They evaluated a dataset of 122 logical vulnerabilities and a framework to evaluate patches for logical weaknesses. |
| Outcome: | The proposed framework evaluates both traditional and LLM-based approaches for addressing real-world logical vulnerabilities. |