Poisoning Web-Scale Training Datasets is Practical
2024 · 75 citations
https://doi.org/10.1109/sp54263.2024.00179
D
Daniel Paleka
Claim or correct this author profile
Claim this profile or suggest corrections to the name, affiliation, bio, photo, or paper titles. Approved changes appear as verified CitedEvidence overlays.
18
Papers
178
Citations
8
h-index
5
i10-index
Daniel Paleka is an academic researcher. The author has contributed to research in topics: Topic Modeling & Natural Language Processing Techniques & Adversarial Robustness in Machine Learning. The author has an h-index of 8, co-authored 16 publications.
Papers by this author
worksRefusal in Language Models Is Mediated by a Single Direction
2024 · 8 citations
https://doi.org/10.52202/079017-4322
Međunarodna matematička olimpijada 2016. g.
2016 · University of Zagreb University Computing Centre (SRCE) · 0 citations
Large-scale online deanonymization with LLMs
2026 · arXiv (Cornell University) · 0 citations
https://doi.org/10.48550/arxiv.2602.16800
Large-scale online deanonymization with LLMs
2026 · arXiv (Cornell University) · 0 citations
Research topics
topicsChat about Author
Papers
Chat
Click to start Chat