Upload folder using huggingface_hub

README.md

@@ -0,0 +1,13 @@
+---
+license: mit
+language: en
+tags:
+  - rag
+  - cs-552
+  - question-answering
+  - transformer
+  - milestone2
+---
+
+# MNLP_M2_rag_documents
+This is a sample set of documents for use in Retrieval-Augmented Generation (RAG) evaluation.

dataset_info.json

@@ -0,0 +1,11 @@
+{
+  "dataset_name": "MNLP_M2_rag_documents",
+  "description": "Documents used for RAG model retrieval in CS-552 Milestone 2.",
+  "features": {
+    "doc_id": "string",
+    "text": "string",
+    "dataset": "string"
+  },
+  "license": "mit",
+  "task": "retrieval"
+}

documents.jsonl

@@ -0,0 +1,3 @@
+{"doc_id": "doc1", "text": "The Transformer architecture introduced self-attention to model dependencies without regard to distance.", "dataset": "wiki"}
+{"doc_id": "doc2", "text": "Retrieval-Augmented Generation (RAG) enhances generation by retrieving documents relevant to the input query.", "dataset": "manual_curation"}
+{"doc_id": "doc3", "text": "BERT is a transformer model pre-trained on a large corpus and fine-tuned for specific NLP tasks.", "dataset": "bert_paper"}