SlavicNLP-2025-Baselines
Collection
2 items • Updated
# Load model directly
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("SlavicNLP/SlavRoBERTa-PT-Classification-Baseline")
model = AutoModelForSequenceClassification.from_pretrained("SlavicNLP/SlavRoBERTa-PT-Classification-Baseline")Quick Links
Model Card
Baseline model for persuasion techniques multi-class multi-label classification task at SlavicNLP 2025. It is a fine-tuned XLM-RoBERTa-base to the task of persuasion techniques multi-class multi-label classification on Slavic languages and datasets provided on the SlavicNLP 2025 Workshop co-located with ACL 2025
The model was fine-tuned to recognize the following persuasion techniques:
- Appeal to Authority
- Appeal to Fear / Prejudice
- Appeal to Hypocrisy
- Appeal to Pity
- Appeal to Popularity
- Appeal to Time
- Appeal to Values
- Causal Oversimplification
- Consequential Oversimplification
- Conversation Killer
- Doubt
- Exaggeration / Minimisation
- False Dilemma / No Choice
- False Equivalence
- Flag Waving
- Guilt by Association
- Loaded Language
- Name Calling / Labeling
- Obfuscation / Vagueness / Confusion
- Questioning the Reputation
- Red Herring
- Repetition
- Slogans
- Straw Man
- Whataboutism
- Downloads last month
- 4
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="SlavicNLP/SlavRoBERTa-PT-Classification-Baseline")