Training language models to follow instructions with human feedback
Human demonstrations and preference feedback are used to improve how a language model follows instructions.
THE RESEARCH ATLAS
Investigating unintended model actions in our evaluations and internal use
A report of unintended actions observed during model evaluation and internal use, with discussion of safeguards.
Institution and location sources
Location refers to credited institutions or the listed organization, not author citizenship or an exhaustive collaboration list.
Why read this
Read failure cases alongside success benchmarks to understand operational boundaries.
Citation
@misc{anthropic2026investigating,
title = {Investigating unintended model actions in our evaluations and internal use},
author = {Anthropic},
year = {2026},
url = {https://www.anthropic.com/research/investigating-unintended-model-actions}
}Rights remain with the original authors. Refer to the original source for citations and licensing.
Related research
Constitutional AI: Harmlessness from AI Feedback
A set of written principles guides critique, revision and AI-generated preference feedback during alignment.
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
A preference-based objective optimizes a language model directly, avoiding a separate reward-model training stage.