MathTutorBench: A Benchmark for Measuring Open-ended Pedagogical Capabilities of LLM Tutors
Paper • 2502.18940 • Published • 3
Text Classification
Transformers
Safetensors
qwen2
reward
model
education
pedagogical
scaffolding
text-embeddings-inference
# Load model directly
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("eth-nlped/Qwen2.5-1.5B-pedagogical-rewardmodel")
model = AutoModelForSequenceClassification.from_pretrained("eth-nlped/Qwen2.5-1.5B-pedagogical-rewardmodel")Quick Links
The model was trained on paired preferences from the MathDial and MRBench datasets.
To find more information and to cite, see:
@article{macina2025mathtutorbench,
title={MathTutorBench: A Benchmark for Measuring Open-ended\\ Pedagogical Capabilities of LLM Tutors},
author={Jakub Macina, Nico Daheim, Ido Hakimi, Manu Kapur, Iryna Gurevych, Mrinmaya Sachan},
year={2025},
eprint={2502.18940},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2502.18940},
}
- Downloads last month
- 948
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="eth-nlped/Qwen2.5-1.5B-pedagogical-rewardmodel")