Mohammad Mohammadi Amiri
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
ML×1AI×1Comp. Eng.×1NLP×1Crypto×1
Frequent co-authors
Research Timeline
2026
RefusalGuard: Geometry-Preserving Fine-Tuning for Safety in LLMs
The paper introduces RefusalGuard, a novel fine-tuning framework that preserves the geometric structure of safety-relevant representations in LLMs, thereby mitigating the degradation of refusal behavior during task-specific adaptation.
Highlighted terms show continued research focus across papers