This is a transformer-based model trained on the TalkMoves dataset, which is grounded in accountable talk theory and captures features of teacher–student interaction in instructional settings. The model was trained on the teacher component of the dataset to identify discourse patterns associated with pedagogical quality.

The model was developed for the analyses presented in this paper.

Citation

If you find this model useful for your work, please cite:

@article{abdulsalam2025large,
  title={Large Language Models Approach Expert Pedagogical Quality in Math Tutoring but Differ in Instructional and Linguistic Profiles},
  author={Abdulsalam, Ramatu Oiza and Aroyehun, Segun},
  journal={arXiv preprint arXiv:2512.20780},
  year={2025}
}
Downloads last month
57
Safetensors
Model size
0.4B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Paper for saroyehun/Talkmove-bert