24 citations · 32 across the 6 of their papers we have counts for
7 papers
Programming by Backprop: An Instruction is Worth 100 Examples When Finetuning LLMs
Jonathan Cook, Silvia Sapora, Arash Ahmadian +4
Large language models (LLMs) are typically trained to acquire behaviours from demonstrations or experience, yet much of their training data is declarative: instructions, rules, and…
Ctrl-Z: Controlling AI Agents via Resampling
Aryan Bhatt, Cody Rushing, Adam Kaufman +5
Control evaluations measure whether monitoring and security protocols for AI systems prevent intentionally subversive AI models from causing harm. Our work presents the first contr…
Auditing language models for hidden objectives
Samuel Marks, Johannes Treutlein, Trenton Bricken +32
We study the feasibility of conducting alignment audits: investigations into whether models have undesired objectives. As a testbed, we train a language model with a hidden objecti…
Factorio Learning Environment
Jack Hopkins, Mart Bakler, Akbir Khan
Large Language Models (LLMs) are rapidly saturating existing benchmarks, necessitating new open-ended evaluations. We introduce the Factorio Learning Environment (FLE), based on th…
Alignment faking in large language models
Ryan Greenblatt, Carson Denison, Benjamin Wright +17
We present a demonstration of a large language model engaging in alignment faking: selectively complying with its training objective in training to prevent modification of its beha…
Adaptive Deployment of Untrusted LLMs Reduces Distributed Threats
Jiaxin Wen, Vivek Hebbar, Caleb Larson +9
As large language models (LLMs) become increasingly capable, it is prudent to assess whether safety measures remain effective even if LLMs intentionally try to bypass them. Previou…
Language Models Learn to Mislead Humans via RLHF
Jiaxin Wen, Ruiqi Zhong, Akbir Khan +6
Language models (LMs) can produce errors that are hard to detect for humans, especially when the task is complex. RLHF, the most popular post-training method, may exacerbate this p…