Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming
Paper
•
2501.18837
•
Published
•
8
Scott Goodfriend
sgoodfriend
AI & ML interests
None yet
Recent Activity
authored
a paper
7 days ago
Constitutional Classifiers: Defending against Universal Jailbreaks
across Thousands of Hours of Red Teaming
updated
a model
almost 2 years ago
sgoodfriend/ppo-Microrts-selfplay-unet-decay
updated
a model
almost 2 years ago
sgoodfriend/ppo-Microrts-selfplay-unet
Organizations
None yet
sgoodfriend's activity
authored
a
paper
7 days ago
updated
19
models
almost 2 years ago
sgoodfriend/ppo-Microrts-selfplay-unet-decay
Reinforcement Learning
•
Updated
•
2
sgoodfriend/ppo-Microrts-selfplay-unet
Reinforcement Learning
•
Updated
sgoodfriend/ppo-CartPole-v1
Reinforcement Learning
•
Updated
sgoodfriend/ppo-MountainCar-v0
Reinforcement Learning
•
Updated
sgoodfriend/ppo-Acrobot-v1
Reinforcement Learning
•
Updated
sgoodfriend/ppo-LunarLander-v2
Reinforcement Learning
•
Updated
sgoodfriend/ppo-HalfCheetahBulletEnv-v0
Reinforcement Learning
•
Updated
sgoodfriend/ppo-MountainCarContinuous-v0
Reinforcement Learning
•
Updated
sgoodfriend/ppo-BipedalWalker-v3
Reinforcement Learning
•
Updated
sgoodfriend/ppo-AntBulletEnv-v0
Reinforcement Learning
•
Updated
sgoodfriend/ppo-CarRacing-v0
Reinforcement Learning
•
Updated
sgoodfriend/ppo-HopperBulletEnv-v0
Reinforcement Learning
•
Updated
sgoodfriend/ppo-Walker2DBulletEnv-v0
Reinforcement Learning
•
Updated
sgoodfriend/ppo-PongNoFrameskip-v4
Reinforcement Learning
•
Updated
sgoodfriend/ppo-BreakoutNoFrameskip-v4
Reinforcement Learning
•
Updated
sgoodfriend/ppo-SpaceInvadersNoFrameskip-v4
Reinforcement Learning
•
Updated
sgoodfriend/ppo-QbertNoFrameskip-v4
Reinforcement Learning
•
Updated
sgoodfriend/a2c-MountainCar-v0
Reinforcement Learning
•
Updated
sgoodfriend/a2c-CartPole-v1
Reinforcement Learning
•
Updated