crisistransformers/tiny

CrisisTransformers Mini Models

This is the Tiny variant among the mini models (Medium, Small, Tiny) that were introduced in the paper "Actionable Help" in Crises: A Novel Dataset and Resource-Efficient Models for Identifying Request and Offer Social Media Posts.

These models are the first crisis-specific mini models optimized for deployment in resource-constrained settings. Across 13 crisis classification tasks, our mini models surpass BERT (also outperform or match the performance of RoBERTa, MPNet, and BERTweet), offering higher accuracy with significantly smaller sizes and faster speeds. The Medium model is 47% smaller with 3.8% higher accuracy at 3.5x speed, the Small model is 68% smaller with a 1.8% accuracy gain at 7.7x speed, and the Tiny model, 83% smaller, matches BERT's accuracy at 18.6x speed. All models outperform existing distilled variants, setting new benchmarks. Refer to the associated paper for more details.

Architecture

mini model	# attention heads	# layers	intermediate size	output size	# parameters	source
Medium	8	8	2048	512	58 million	crisistransformers/medium
Small	6	6	1536	384	35 million	crisistransformers/small
Tiny	4	4	1024	256	19 million	crisistransformers/tiny

Uses

These models should be finetuned for downstream tasks just like BERT and RoBERTa.

Citation

If you use these models in your research/project, please cite the following paper:

@article{lamsal2025actionable,
      title={"Actionable Help" in Crises: A Novel Dataset and Resource-Efficient Models for Identifying Request and Offer Social Media Posts}, 
      author={Rabindra Lamsal and
              Maria Rodriguez Read and
              Shanika Karunasekera and
              Muhammad Imran},
      year={2025},
      eprint={2502.16839},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}