Anushhh/KannadaPromptBench
KannadaPromptBench A benchmark dataset for evaluating prompt strategy sensitivity in Kannada, a low-resource Dravidian language. Dataset Summary Language: Kannada (kn) Tasks: Sentiment Analysis (100), Question Answering (75), Summarization (50) Total: 225 culturally grounded samples Inter-annotator agreement: Cohen's κ > 0.80 Dataset Structure Each sample contains: id, task, input_text, label, difficulty, domain. Citation Please… See the full description on the dataset page: https://huggingface.co/datasets/Anushhh/KannadaPromptBench.
Upload experiment_log.json with huggingface_hub
Upload annotation/iaa_samples.csv with huggingface_hub
Upload analysis/adversarial_samples.csv with huggingface_hub
Upload analysis/adversarial_eval_results.csv with huggingface_hub
Upload analysis/statistical_significance.csv with huggingface_hub
Upload analysis/error_analysis.csv with huggingface_hub
Upload results/openrouter_results.csv with huggingface_hub
Upload results/sarvam_results.csv with huggingface_hub
Upload results/cerebras_results.csv with huggingface_hub
Upload results/kannada_results.csv with huggingface_hub
Upload figures/fig3_cot_gain.png with huggingface_hub
Upload figures/fig2_heatmap.png with huggingface_hub
Upload figures/fig1_accuracy_by_task.png with huggingface_hub
Upload data/kannada_promptbench.json with huggingface_hub
Upload README.md with huggingface_hub
initial commit