ADCO AI Labs
ADCO AI Labs is our artificial intelligence research lab. We study data, model evaluation and the relationship between people and AI systems.
Publications
Analysis from ADCO AI Labs on datasets, evaluation and human feedback.
A useful dataset is a maintained system.
The value of a dataset depends on what a team can explain, reproduce and improve after the first delivery.
Evaluation starts with the decision.
A score becomes useful when it is tied to a specific workflow, a clear acceptance rule and the evidence behind each judgment.
Human feedback needs a disagreement policy.
The useful signal is often in the reasons people disagree. A review process needs a way to preserve and resolve those differences.
What we study
Our research interests include the data AI learns from, the methods used to evaluate its behavior, and the role of human judgment in AI systems.
We examine how dataset quality shapes learning, what an evaluation can tell us about a model, and how human feedback can support better decisions. Our publications share analysis of these questions with links to the research behind them.