1 code implementation • 15 Apr 2024 • Usman Anwar, Abulhair Saparov, Javier Rando, Daniel Paleka, Miles Turpin, Peter Hase, Ekdeep Singh Lubana, Erik Jenner, Stephen Casper, Oliver Sourbut, Benjamin L. Edelman, Zhaowei Zhang, Mario Günther, Anton Korinek, Jose Hernandez-Orallo, Lewis Hammond, Eric Bigelow, Alexander Pan, Lauro Langosco, Tomasz Korbak, Heidi Zhang, Ruiqi Zhong, Seán Ó hÉigeartaigh, Gabriel Recchia, Giulio Corsi, Alan Chan, Markus Anderljung, Lilian Edwards, Yoshua Bengio, Danqi Chen, Samuel Albanie, Tegan Maharaj, Jakob Foerster, Florian Tramer, He He, Atoosa Kasirzadeh, Yejin Choi, David Krueger
This work identifies 18 foundational challenges in assuring the alignment and safety of large language models (LLMs).
no code implementations • 25 Jul 2023 • Tomohiro Sawada, Daniel Paleka, Alexander Havrilla, Pranav Tadepalli, Paula Vidas, Alexander Kranias, John J. Nay, Kshitij Gupta, Aran Komatsuzaki
As a subset of ARB, we introduce a challenging set of math and physics problems which require advanced symbolic reasoning and domain knowledge.
2 code implementations • 16 Jun 2023 • Lukas Fluri, Daniel Paleka, Florian Tramèr
If machine learning models were to achieve superhuman abilities at various reasoning or decision-making tasks, how would we go about evaluating such models, given that humans would necessarily be poor proxies for ground truth?
no code implementations • 20 Feb 2023 • Nicholas Carlini, Matthew Jagielski, Christopher A. Choquette-Choo, Daniel Paleka, Will Pearce, Hyrum Anderson, Andreas Terzis, Kurt Thomas, Florian Tramèr
Deep learning models are often trained on distributed, web-scale datasets crawled from the internet.
no code implementations • 3 Oct 2022 • Javier Rando, Daniel Paleka, David Lindner, Lennart Heim, Florian Tramèr
We then reverse-engineer the filter and find that while it aims to prevent sexual content, it ignores violence, gore, and other similarly disturbing content.
no code implementations • 8 Jul 2022 • Daniel Paleka, Amartya Sanyal
In supervised learning, it has been shown that label noise in the data can be interpolated without penalties on test accuracy.