Update README.md

README.md

@@ -20,7 +20,8 @@ configs:
 ---
 # Accel-IR Gold Standard Dataset
 
-This dataset contains expert-annotated Question-Answer pairs for the Particle Accelerator Domain, as described in the Master's Thesis *"From Dataset to Optimization"* by Qing Dai.
+This dataset contains 390 pure expert-annotated Question-Chunk pairs for the **Particle Accelerator Domain**, as described in the Master's Thesis *"From Dataset to Optimization: A Benchmarking Framework for Information Retrieval in the Particle Accelerator Domain"* by Qing Dai.
+With expert-annotated relevance between the question and chunk, this serves as gold standard for information retrieval models evaluation in the particle accelerator domain.
 
 ## Dataset Structure