Get started
Author

Ahmed Salem

Observed name: Salem, Ahmed
Observed institution: Microsoft (United States)
OpenAlex ID: A5125404114
The Hawthorne Effect in Reasoning Models: Evaluating and Steering Test Awareness
2025 · Sahar Abdelnabi, Ahmed Salem · neural information processing systems · 2
Power bank rated capacity vs watt-hour energy and airline carry-on category
2026 · Salem, Ahmed · Mendeley Data
Divide, Consult, Conquer: Capability Laundering Through Aligned LLMs
2026 · Russinovich, Mark, Bullwinkel, Blake, et al. · arXiv (Cornell University)
PDF
SpecGuard: Inference-Time Backdoor Detection For Free
2026 · Wen, Rui, Salem, Ahmed, et al. · arXiv (Cornell University)
PDF
Phantom References: Hallucinated Citations That Survive Peer Review at Top-Tier Conferences
2026 · Russinovich, Mark, Kumar, Ram Shankar Siva, et al. · arXiv (Cornell University)
PDF
Anker USB-C Charger and Cable Compatibility Dataset for E-Commerce Verification
2026 · Salem, Ahmed · Mendeley Data
MetaBackdoor: Exploiting Positional Encoding as a Backdoor Attack Surface in LLMs
2026 · Wen, Rui, Russinovich, Mark, et al. · arXiv (Cornell University)
PDF
QSTN: A Modular Framework for Robust Questionnaire Inference with Large Language Models
2026 · Association for Computational Linguistics 2026, Ahnert, Georg, et al. · Underline Science Inc.
ConVerse: Benchmarking Contextual Safety in Agent-to-Agent Conversations
2026 · Association for Computational Linguistics 2026, Abdelnabi, Sahar, et al. · Underline Science Inc.
Stateless Yet Not Forgetful: Implicit Memory as a Hidden Channel in LLMs
2026 · Salem, Ahmed, Paverd, Andrew, et al. · arXiv (Cornell University)
PDF
GRP-Obliteration: Unaligning LLMs With a Single Unlabeled Prompt
2026 · Russinovich, Mark, Cai, Yanan, et al. · arXiv (Cornell University)
PDF
Generative AI-Driven Data Transformation for Enhanced Machine Learning Performance (Student Abstract)
2026 · Association for Artificial Intelligence 2026, Jim, Carol, et al. · Underline Science Inc.
ConVerse: Benchmarking Contextual Safety in Agent-to-Agent Conversations
2026 · Amr Gomaa, Ahmed Salem, et al. · Findings of the Association for Computational Linguistics: EACL
Detection and Attribution of Models Trained on Generated Data
2024 · Han, Ge, Salem, Ahmed, et al. · Publications of the UdS (Saarland University)
Get a Model! Model Hijacking Attack Against Machine Learning Models
2022 · Salem, Ahmed, Backes, Michael, et al. · Publications of the UdS (Saarland University)
BadNL: Backdoor Attacks against NLP Models with Semantic-preserving Improvements
2021 · Chen, Xiaoyi, Salem, Ahmed, et al. · Publications of the UdS (Saarland University)
Works count: 16
Citation count: 2
H-index: 1
I10-index:
Year
Topic
Type