Skip to content

sisl/lohfmh-dataset

Repository files navigation

Expert Evaluation and the Limits of Human Feedback in Mental Health AI Safety Testing Dataset

OpenReview arXiv


This repository contains the dataset and code for the paper "Expert Evaluation and the Limits of Human Feedback in Mental Health AI Safety Testing" published in the ACM Conference on Fairness, Accountability, and Transparency (FAccT) 2026. The dataset contains the human-generated prompts, model-generated responses, and expert evaluations used in the study.

About

Expert Evaluation and the Limits of Human Feedback in Mental Health AI Safety Testing Dataset

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors