Update README.md

README.md

@@ -4,12 +4,12 @@ emoji: 🔍
 colorFrom: blue
 colorTo: green
 sdk: gradio
-sdk_version: 5.44.1
+sdk_version: "4.31.5"
 app_file: app.py
 pinned: false
 ---
 
-# SBERT + FAISS Semantic Search
+# SBERT + FAISS Semantic Search + Evaluation Metrics
 
 This Hugging Face Space hosts a **semantic search system** built with:
 
@@ -22,7 +22,7 @@ This Hugging Face Space hosts a **semantic search system** built with:
 
 ## 🔹 Features
 - Enter a **query** to retrieve the **Top-10 most similar passages**.  
-- Optionally provide **ground truth relevant passages** (one per line) to compute **IR metrics**:  
+- Computes **true IR metrics** when the query matches one in MS MARCO validation set:
   - Precision@10  
   - Recall@10  
   - F1-score  
@@ -33,11 +33,10 @@ This Hugging Face Space hosts a **semantic search system** built with:
 
 ## 🔹 How to Use
 1. Type a query into the input box.  
-2. (Optional) Paste one or more relevant passages into the second box, each on a new line.  
-3. Press **Submit**.  
-4. View:  
-   - **Top-10 retrieved passages** with FAISS similarity scores  
-   - **Evaluation metrics** if ground truth passages were provided  
+2. Press **Submit**.  
+3. View:  
+   - **Top-10 retrieved passages** with similarity scores  
+   - **Evaluation metrics** if the query exists in the validation set  
 
 ---
 
@@ -58,4 +57,4 @@ If you use this system in research, please cite:
 ---
 
 ## 🔹 Author
-Built for a research project on **user-centered evaluation of semantic search systems**.
+Built for a research project on **user-centered evaluation of semantic search systems**.