EMNLP 2026 · MainFirst author
MedRoundsQA: A Persona and Difficulty Aware Evaluation for Multi-Turn Medical Consultations
Youssef Mohamed, Ahmed Heakl, Qinrong Cui, Junhong Liang, Rafiq Ali, Bdour Babillie, Nazira Dunbayeva, Lang Gao, Omar Hussein, Ahmed Nada, Ahmed Mohamed Magdy Mohamed, Jinghui Liu, Salman Khan, Imran Razzak, Yuxia Wang, Xiuying Chen
Same case, same facts, different patient: moving from single-turn diagnosis to a real consultation costs 15 LLMs 13–39 points, and how educated the patient sounds shifts accuracy by another 7–8.
PaperCode & data
ICML 2026
The Cylindrical Representation Hypothesis for Language Model Steering
Lang Gao, Jinghui Zhang, Wei Liu, Fengxian Ji, Chenxi Wang, Zirui Song, Akash Ghosh, Youssef Mohamed, Preslav Nakov, Xiuying Chen
Relaxing the orthogonality assumption of the Linear Representation Hypothesis explains why steering directions interact and destabilise model behaviour.
PaperCode
SecureComm 2025
Two Birds with One Stone: Multi-Task Detection and Attribution of LLM-Generated Text
Zixin Rao*, Youssef Mohamed*, Shang Liu, Zeyan Liu (*equal contribution)
A multitask model that both detects machine-generated text and attributes it to its source LLM, across languages.
Paper
ACLing 2024
ResuméAtlas: Revisiting Resume Classification with Large-Scale Datasets and Large Language Models
Ahmed Heakl*, Youssef Mohamed*, Noran Mohamed*, Aly Elsharkawy, Ahmed Zaky (*equal contribution)
PaperCodeDataset