Text Classification
Transformers
Safetensors
English
deberta-v2
citation-verification
retrieval-augmented-generation
rag
cross-lingual
deberta
cross-encoder
nli
attribution
Eval Results (legacy)
text-embeddings-inference
Instructions to use convexray/alignment-module-cross-encoder-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use convexray/alignment-module-cross-encoder-base with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="convexray/alignment-module-cross-encoder-base")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("convexray/alignment-module-cross-encoder-base") model = AutoModelForSequenceClassification.from_pretrained("convexray/alignment-module-cross-encoder-base", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit History
14 days ago
-
Update README.md
Verified signaturecf58f4c
-
Delete model_card.md
Verified signaturea38b959 -
Update README.md
Verified signaturec4c1f5b -
Create README.md
Verified signaturedb7bdf2 -
create model card
Verified signature--- language: - en license: mit library_name: transformers tags: - citation-verification - retrieval-augmented-generation - rag - cross-lingual - deberta - cross-encoder - nli - attribution pipeline_tag: text-classification datasets: - fever - din0s/asqa - miracl/hagrid metrics: - f1 - precision - recall - accuracy - roc_auc base_model: microsoft/deberta-v3-base model-index: - name: dualtrack-alignment-module results: - task: type: text-classification name: Citation Verification metrics: - type: f1 value: 0.89 name: F1 Score - type: accuracy value: 0.87 name: Accuracy - type: roc_auc value: 0.94 name: ROC-AUC --- # DualTrack Alignment Module > **Anonymous submission to ACL 2026** A cross-encoder model for detecting **citation drift** in Retrieval-Augmented Generation (RAG) systems. Given a user-facing claim, an evidence representation, and a source passage, the model predicts whether the citation is valid (the sourc…ee26f70 -
Rename README.md to model_card.md
Verified signature--- language: - en license: mit library_name: transformers tags: - citation-verification - retrieval-augmented-generation - rag - cross-lingual - deberta - cross-encoder - nli - attribution pipeline_tag: text-classification datasets: - fever - din0s/asqa - miracl/hagrid metrics: - f1 - precision - recall - accuracy - roc_auc base_model: microsoft/deberta-v3-base model-index: - name: dualtrack-alignment-module results: - task: type: text-classification name: Citation Verification metrics: - type: f1 value: 0.89 name: F1 Score - type: accuracy value: 0.87 name: Accuracy - type: roc_auc value: 0.94 name: ROC-AUC --- # DualTrack Alignment Module > **Anonymous submission to ACL 2026** A cross-encoder model for detecting **citation drift** in Retrieval-Augmented Generation (RAG) systems. Given a user-facing claim, an evidence representation, and a source passage, the model predicts whether the citation is valid (the sourc…bac405a -
Update README.md
Verified signature--- language: - en license: mit library_name: transformers tags: - citation-verification - retrieval-augmented-generation - rag - cross-lingual - deberta - cross-encoder - nli - attribution pipeline_tag: text-classification datasets: - fever - din0s/asqa - miracl/hagrid metrics: - f1 - precision - recall - accuracy - roc_auc base_model: microsoft/deberta-v3-base model-index: - name: dualtrack-alignment-module results: - task: type: text-classification name: Citation Verification metrics: - type: f1 value: 0.89 name: F1 Score - type: accuracy value: 0.87 name: Accuracy - type: roc_auc value: 0.94 name: ROC-AUC --- # DualTrack Alignment Module > **Anonymous submission to ACL 2026** A cross-encoder model for detecting **citation drift** in Retrieval-Augmented Generation (RAG) systems. Given a user-facing claim, an evidence representation, and a source passage, the model predicts whether the citation is valid (the sourc…5a5e091 -
Upload folder using huggingface_hub
Verified signature32f322b -
initial commit
Verified signature3087c40