Search Results for author:
Found 12 papers, 6 papers with code
Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
We find that such backdoor behavior can be made persistent, so that it is not removed by standard safety training techniques, including supervised fine-tuning, reinforcement learning, and adversarial training (eliciting unsafe behavior and then training to remove it).
Managing AI Risks in an Era of Rapid Progress
In this short consensus paper, we outline risks from upcoming, advanced AI systems.
Specific versus General Principles for Constitutional AI
Constitutional AI offers an alternative, replacing human feedback with feedback from AI models conditioned only on a list of written principles.
How to Catch an AI Liar: Lie Detection in Black-Box LLMs by Asking Unrelated Questions
Large language models (LLMs) can "lie", which we define as outputting false statements despite "knowing" the truth in a demonstrable sense.
The Alignment Problem from a Deep Learning Perspective
In coming years or decades, artificial general intelligence (AGI) may surpass human capabilities at many critical tasks.
Prioritized Training on Points that are Learnable, Worth Learning, and Not Yet Learnt
But most computation and time is wasted on redundant and noisy points that are already learnt or not learnable.
Prioritized training on points that are learnable, worth learning, and not yet learned (workshop version)
We introduce Goldilocks Selection, a technique for faster model training which selects a sequence of training points that are "just right".
Quantifying Ignorance in Individual-Level Causal-Effect Estimates under Hidden Confounding
We study the problem of learning conditional average treatment effects (CATE) from high-dimensional, observational data with unobserved confounders.
How Robust are the Estimated Effects of Nonpharmaceutical Interventions against COVID-19?
To what extent are effectiveness estimates of nonpharmaceutical interventions (NPIs) against COVID-19 influenced by the assumptions our models make?
Identifying Causal-Effect Inference Failure with Uncertainty-Aware Models
We show that our methods enable us to deal gracefully with situations of "no-overlap", common in high-dimensional data, where standard applications of causal effect approaches fail.
Active Inverse Reward Design
We propose structuring this process as a series of queries asking the user to compare between different reward functions.
Occam's razor is insufficient to infer the preferences of irrational agents
Inverse reinforcement learning (IRL) attempts to infer human rewards or preferences from observed behavior.
