Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
json
Languages:
Vietnamese
Size:
10K - 100K
License:
Commit History
-
Upload README.md with huggingface_hub
Verified signature0fec8d6
-
Upload README.md with huggingface_hub
Verified signature588b219
-
Upload README.md with huggingface_hub
Verified signature6b49b61 -
Create README.md
Verified signature# Dataset Card for “ViANLI” ## Dataset Summary **ViANLI (Vietnamese Adversarial Natural Language Inference)** is the first adversarial benchmark dataset for Vietnamese NLI, designed to evaluate model robustness against complex linguistic phenomena. The dataset was constructed using a **human-and-machine-in-the-loop approach** with **multi-round adversarial generation** and **dual human–machine verification**. ViANLI contains over **10,000 high-quality premise–hypothesis pairs** across 13 diverse domains from Vietnamese news articles. Each pair is labeled as **entailment**, **contradiction**, or **neutral**, following the standard NLI framework. The dataset provides a challenging benchmark for evaluating reasoning robustness and supports research in both **Vietnamese** and **multilingual NLI**. --- ## Languages - **Vietnamese (vi)** --- ## Dataset Structure ### Data Instances Each instance in ViANLI is a JSON line with the following fields: | Field | Type | Description |…787104a -
Upload vianli_dev.jsonl with huggingface_hub
Verified signature571ed41 -
Upload vianli_test.jsonl with huggingface_hub
Verified signatured783512 -
Upload vianli_train.jsonl with huggingface_hub
Verified signatureb65d6d8 -
initial commit
Verified signature50ffc20