Paul Kassianik
5 papers · 2023–2025 · 5 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+4 more ↓ Show less ↑
π Interdisciplinary Bridge π Renaissance Researcher (5) πΊοΈ Taxonomy Completionist (13) π§ Keyword Pioneer π Conference Polyglot (5)
π£
Hot Topic Early Bird
π
Cross-Pollinator
(9)
π₯
Mega-Team
(23)
π
Trend Setter
Conferences
AACL (1)
ICML (1)
IJCNLP (1)
JMLR (1)
NIPS (1)
Top co-authors
Keywords
linear separability
(2)
latent space geometry
(2)
prompt engineering
(1)
model safety
(1)
semantic representation
(1)
ai safety
(1)
distributed computing
(1)
safety alignment
(1)
model alignment
(1)
time series forecasting
(1)
hyperparameter tuning
(1)
adversarial attack
(1)
black-box attack
(1)
language model
(1)
model ensemble
(1)
jailbreaking attack
(1)
jailbreak attack
(1)
model ensembling
(1)
red teaming
(1)
adversarial prompt generation
(1)
Papers
Large Language Models Encode Semantics and Alignment in Linearly Separable Representations
AACL 2025
Adversarial Reasoning at Jailbreaking Time
ICML 2025
Large Language Models Encode Semantics and Alignment in Linearly Separable Representations
IJCNLP 2025
Tree of Attacks: Jailbreaking Black-Box LLMs Automatically
NIPS 2024
Merlion: End-to-End Machine Learning for Time Series
JMLR 2023