davidfearne
commited on
Update app.py
Browse files
app.py
CHANGED
@@ -88,7 +88,12 @@ st.sidebar.caption(f"Session ID: {genuuid()}")
|
|
88 |
|
89 |
|
90 |
# Main chat interface
|
91 |
-
st.
|
|
|
|
|
|
|
|
|
|
|
92 |
|
93 |
# User ID Input
|
94 |
user_id = st.text_input("Experiment ID:", key="user_id")
|
|
|
88 |
|
89 |
|
90 |
# Main chat interface
|
91 |
+
st.markdown("""Query translation in a Retrieval-Augmented Generation (RAG) architecture is the process where an LLM acts as a translator between the user's natural language input and the underlying retrieval system. When a user sends a message, the query translator takes this input and:
|
92 |
+
|
93 |
+
** Adds Context: Enriches the user's input with relevant context (e.g., expanding vague questions or specifying details) to make it more precise.
|
94 |
+
** Converts to Concise Query: Reformulates the input into a succinct and effective query optimized for the retrieval system's semantic search capabilities.
|
95 |
+
|
96 |
+
This ensures that the retrieval system gets a clear and focused query, increasing the relevance of the information it retrieves. The query translator essentially bridges the gap between human conversational language and the technical requirements of a semantic retrieval system.""")
|
97 |
|
98 |
# User ID Input
|
99 |
user_id = st.text_input("Experiment ID:", key="user_id")
|