CoUDA: Coherence Evaluation via Unified Data Augmentation
Paper • 2404.00681 • Published
How to use BVendeville/couda-albert-xxl with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-classification", model="BVendeville/couda-albert-xxl") # Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("BVendeville/couda-albert-xxl", dtype="auto")Fine-tuned ALBERT-xxl model for the CoUDA (Coherence via Unified Discourse Analysis) coherence metric, used in the simpsy text simplification evaluation library.
from simpsy import evaluate
df = evaluate(
["The cat sat on the mat. It watched the mouse. Then it pounced."],
families=["couda"],
)
Or point simpsy at this repo explicitly:
export SIMPSY_COUDA_HF_REPO=BVendeville/couda-albert-xxl
Zhang et al. (2024), "CoUDA: Coherence Evaluation via Unified Data Augmentation" — https://arxiv.org/abs/2404.00681.