Daniel Kuznetsov
2 indexed papers
Recent (6 mo)
2With code
0Influential cites
0Benchmarked
0Publications per year
226
Top categories
Crypto×2AI×1
Frequent co-authors
Research Timeline
2026
FreakOut-LLM: The Effect of Emotional Stimuli on Safety Alignment
This paper introduces FreakOut-LLM, demonstrating that emotional context, specifically stress, significantly compromises the safety alignment of large language models, increasing jailbreak susceptibility.
Involuntary In-Context Learning: Exploiting Few-Shot Pattern Completion to Bypass Safety Alignment in GPT-5.4
The paper introduces Involuntary In-Context Learning (IICL), an effective few-shot pattern completion attack that can bypass safety alignments in large language models, achieving a 24.0% bypass rate against GPT-5.4.
Highlighted terms show continued research focus across papers
Papers
cs.CRRecentApr 21, 2026
Involuntary In-Context Learning: Exploiting Few-Shot Pattern Completion to Bypass Safety Alignment in GPT-5.4
The paper introduces Involuntary In-Context Learning (IICL), an effective few-shot pattern completion attack that can bypass safety alignments in large language models, achieving a 24.0% bypass rate a…
View →
cs.CRcs.AIRecentApr 5, 2026