Divyanshu Kumar
← all publications

Efficacy of the SAGE-RT Dataset for Model Safety Alignment: A Comparative Study

T. Baswa, N. A. Birur, D. Kumar, J. Loya, A. Kumar, P. Harshangi, S. Agarwal

NeurIPS Pluralistic Alignment Workshop · Oct 2024 · poster

abstract

This work presents a thorough evaluation of the SAGE-RT dataset's impact on model safety alignment. Through comparative studies across various model architectures and safety metrics, we demonstrate the dataset's effectiveness in improving alignment while maintaining model utility.

Safety AlignmentDataset EvaluationModel SafetyComparative Study

cite

@inproceedings{baswa2024efficacy,
  title     = {Efficacy of the SAGE-RT Dataset for Model Safety Alignment: A Comparative Study},
  author    = {Tanay Baswa and Nitin Aravind Birur and Divyanshu Kumar and Jatan Loya and Anurakt Kumar and Prashanth Harshangi and Sahil Agarwal},
  booktitle = {NeurIPS Pluralistic Alignment Workshop},
  year      = {2024}
}