Mohammad Hossein Rohban
Broad Institute · US
Relevante Arbeiten
Meistzitierte Publikationen im Bereich Gesundheit & MedTech
Lying to Win: Assessing LLM Deception through Human-AI Games and Parallel-World Probing
2026 · 0 Zit. · Open MIND
The Judge Who Never Admits: Hidden Shortcuts in LLM-based Evaluation
2026 · 0 Zit. · ArXiv.org
Debate as Reward: A Multi-Agent Reward System for Scientific Ideation via RL Post-Training
2026 · 0 Zit. · arXiv (Cornell University)
Lying to Win: Assessing LLM Deception through Human-AI Games and Parallel-World Probing
2026 · 0 Zit. · arXiv (Cornell University)
Debate as Reward: A Multi-Agent Reward System for Scientific Ideation via RL Post-Training
2026 · 0 Zit. · arXiv (Cornell University)
The Judge Who Never Admits: Hidden Shortcuts in LLM-based Evaluation
2026 · 0 Zit. · Open MIND