Search Results for author:
Found 5 papers, 3 papers with code
AI Control: Improving Safety Despite Intentional Subversion
This protocol asks GPT-4 to write code, and then asks another instance of GPT-4 whether the code is backdoored, using various techniques to prevent the GPT-4 instances from colluding.
Preventing Language Models From Hiding Their Reasoning
Large language models (LLMs) often benefit from intermediate steps of reasoning to generate answers to complex problems.
Benchmarks for Detecting Measurement Tampering
When training powerful AI systems to perform complex tasks, it may be challenging to provide training signals which are robust to optimization.
Large Language Models Sometimes Generate Purely Negatively-Reinforced Text
When using adversarial training, it is common practice to train against the most egregious failures.
Language models are better than humans at next-token prediction
Current language models are considered to have sub-human capabilities at natural language tasks like question-answering or writing code.
