Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming

no code implementations • 31 Jan 2025 • Mrinank Sharma, Meg Tong, Jesse Mu, Jerry Wei, Jorrit Kruthoff, Scott Goodfriend, Euan Ong, Alwin Peng, Raj Agarwal, Cem Anil, Amanda Askell, Nathan Bailey, Joe Benton, Emma Bluemke, Samuel R. Bowman, Eric Christiansen, Hoagy Cunningham, Andy Dau, Anjali Gopal, Rob Gilson, Logan Graham, Logan Howard, Nimit Kalra, Taesung Lee, Kevin Lin, Peter Lofgren, Francesco Mosconi, Clare O'Hara, Catherine Olsson, Linda Petrini, Samir Rajani, Nikhil Saxena, Alex Silverstein, Tanya Singh, Theodore Sumers, Leonard Tang, Kevin K. Troy, Constantin Weisser, Ruiqi Zhong, Giulio Zhou, Jan Leike, Jared Kaplan, Ethan Perez

Large language models (LLMs) are vulnerable to universal jailbreaks-prompting strategies that systematically bypass model safeguards and enable users to carry out harmful processes that require many model interactions, like manufacturing illegal substances at scale.

Red Teaming

Fast Bidirectional Probability Estimation in Markov Models

no code implementations • NeurIPS 2015 • Siddhartha Banerjee, Peter Lofgren

We develop a new bidirectional algorithm for estimating Markov chain multi-step transition probabilities: given a Markov chain, we want to estimate the probability of hitting a given target state in $\ell$ steps after starting from a given source distribution.

Personalized PageRank Estimation and Search: A Bidirectional Approach

1 code implementation • 21 Jul 2015 • Peter Lofgren, Siddhartha Banerjee, Ashish Goel

First, for the problem of estimating Personalized PageRank (PPR) from a source distribution to a target node, we present a new bidirectional estimator with simple yet strong guarantees on correctness and performance, and 3x to 8x speedup over existing estimators in experiments on a diverse set of networks.