bert-mini-toxicity

This is a toxicity classifier fine-tuned using the gravitee-io/textdetox-multilingual-toxicity-dataset. The model supports a wide range of languages and is trained for toxicity classification ("not-toxic", "toxic").

We perform an 85/15 train-test split per language based on the textdetox dataset. All credits go to the authors of the original corpora.

Performance Overview

While the model performance differs from gravitee-io/distilbert-multilingual-toxicity-classifier Some languages still make the cut, even with the base model being lightweight and trained on English as per the model card.

Original model

Language	eval F1	train F1	Δ F1
en	0.955823	0.992248	-0.036425
fr	0.898551	0.975587	-0.077036
de	0.873016	0.958611	-0.085595
hi	0.869221	0.923933	-0.054713
ru	0.836565	0.842410	-0.005845
es	0.807245	0.918570	-0.111325
it	0.804124	0.903806	-0.099683
uk	0.815013	0.818527	-0.003514
tt	0.799443	0.840136	-0.040693
ja	0.761658	0.774920	-0.013262
hin	0.740053	0.845366	-0.105313
ar	0.675386	0.710673	-0.035287
am	0.639659	0.688475	-0.048816
zh	0.634211	0.668962	-0.034752
he	0.444444	0.554080	-0.109635

Quantized model (ONNX)

Language	eval F1	train F1	Δ F1
en	0.955705	0.992241	-0.036536
fr	0.899329	0.972551	-0.073222
de	0.875000	0.954798	-0.079798
hi	0.866310	0.922319	-0.056009
ru	0.831933	0.837851	-0.005918
uk	0.801642	0.819167	-0.017525
it	0.801061	0.900024	-0.098963
tt	0.793696	0.838308	-0.044612
es	0.783673	0.908962	-0.125288
ja	0.759358	0.775797	-0.016439
hin	0.723810	0.847162	-0.123352
ar	0.678788	0.706675	-0.027887
am	0.641026	0.688030	-0.047005
zh	0.632844	0.658548	-0.025704
he	0.409357	0.527363	-0.118006

🤗 Usage

from transformers import AutoTokenizer
from optimum.onnxruntime import ORTModelForSequenceClassification
import numpy as np
# Load model and tokenizer using optimum
model = ORTModelForSequenceClassification.from_pretrained(
 "gravitee-io/bert-mini-toxicity",
 file_name="model.quant.onnx"
)
tokenizer = AutoTokenizer.from_pretrained("gravitee-io/bert-mini-toxicity")
# Tokenize input
text = "Your text here"
inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True)
# Run inference
outputs = model(**inputs)
logits = outputs.logits
# Optional: convert to probabilities
probs = 1 / (1 + np.exp(-logits))
print(probs)

Github Repository

You can check details on how the model was fine-tuned and evaluated on the Github Repository

License

This model is licensed under OpenRAIL++

Citation

@misc{bhargava2021generalization,
      title={Generalization in NLI: Ways (Not) To Go Beyond Simple Heuristics}, 
      author={Prajjwal Bhargava and Aleksandr Drozd and Anna Rogers},
      year={2021},
      eprint={2110.01518},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

@article{DBLP:journals/corr/abs-1908-08962,
  author    = {Iulia Turc and
               Ming{-}Wei Chang and
               Kenton Lee and
               Kristina Toutanova},
  title     = {Well-Read Students Learn Better: The Impact of Student Initialization
               on Knowledge Distillation},
  journal   = {CoRR},
  volume    = {abs/1908.08962},
  year      = {2019},
  url       = {http://arxiv.org/abs/1908.08962},
  eprinttype = {arXiv},
  eprint    = {1908.08962},
  timestamp = {Thu, 29 Aug 2019 16:32:34 +0200},
  biburl    = {https://dblp.org/rec/journals/corr/abs-1908-08962.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

@inproceedings{dementieva2024overview,
  title={Overview of the Multilingual Text Detoxification Task at PAN 2024},
  author={Dementieva, Daryna and Moskovskiy, Daniil and Babakov, Nikolay and Ayele, Abinew Ali and Rizwan, Naquee and Schneider, Frolian and Wang, Xintog and Yimam, Seid Muhie and Ustalov, Dmitry and Stakovskii, Elisei and Smirnova, Alisa and Elnagar, Ashraf and Mukherjee, Animesh and Panchenko, Alexander},
  booktitle={Working Notes of CLEF 2024 - Conference and Labs of the Evaluation Forum},
  editor={Guglielmo Faggioli and Nicola Ferro and Petra Galu{{s}}{{c}}{'a}kov{'a} and Alba Garc{'i}a Seco de Herrera},
  year={2024},
  organization={CEUR-WS.org}
}
@inproceedings{dementieva
-etal-2024-toxicity,
  title = "Toxicity Classification in {U}krainian",
  author = "Dementieva, Daryna and Khylenko, Valeriia and Babakov, Nikolay and Groh, Georg",
  booktitle = "Proceedings of the 8th Workshop on Online Abuse and Harms (WOAH 2024)",
  month = jun,
  year = "2024",
  address = "Mexico City, Mexico",
  publisher = "Association for Computational Linguistics",
  url = "https://aclanthology.org/2024.woah-1.19/",
  doi = "10.18653/v1/2024.woah-1.19",
  pages = "244--255"
}
@inproceedings{DBLP:conf/ecir/BevendorffCCDEFFKMMPPRRSSSTUWZ24,
  author = {Janek Bevendorff and et al.},
  title = {Overview of {PAN} 2024: Multi-author Writing Style Analysis, Multilingual Text Detoxification, Oppositional Thinking Analysis, and Generative {AI} Authorship Verification - Extended Abstract},
  booktitle = {ECIR 2024, Glasgow, UK, March 24-28, 2024, Proceedings, Part {VI}},
  series = {Lecture Notes in Computer Science},
  volume = {14613},
  pages = {3--10},
  publisher = {Springer},
  year = {2024},
  doi = {10.1007/978-3-031-56072-9_1}
}