Update app

app.py

@@ -3,16 +3,20 @@ import pandas as pd
 
 st.set_page_config(layout="wide")
 
-with st.sidebar.expander("Explanation", expanded=False):
-    st.markdown("""This demo allows you to explore the data inside [MRPC](https://www.microsoft.com/en-us/download/details.aspx?id=52398),
-    showing how we can use Word Position Deviation (WPD) and Lexical Deviation (LD) to find different types of paraphrases.
-    By using what we observe from the data, we can also correct numerous labelling errors inside MRPC, presenting the a revision of MRPC termed as MRPC-R1.
-    You can see the rejected and corrected paraphrases by changing the **Display Types** option below.
-    This demo accompanies the paper ["Towards Better Characterization of Paraphrases" (ACL 2022)](https://github.com/tlkh/paraphrase-metrics).""")
+with st.sidebar.expander("📍 Explanation", expanded=False):
+    st.markdown("""
+    This demo allows you to explore the data inside the [MRPC](https://www.microsoft.com/en-us/download/details.aspx?id=52398) dataset. 
+    It illustrates how **Word Position Deviation (WPD)** and **Lexical Deviation (LD)** can be used to find different types of [paraphrase pairs](https://direct.mit.edu/coli/article/39/3/463/1434/What-Is-a-Paraphrase) inside MRPC.
 
-with st.sidebar.expander("Dataset Options", expanded=False):
+    By using what we observe from the data, we can also correct numerous labelling errors inside MRPC, presenting the a revision of MRPC termed as **MRPC-R1**.
+    By changing the **Display Types** option below, you can filter the displayed pairs to show pairs that were rejected (label changed from paraphrase to non-paraphrase) or corrected (inconsistencies corrected).
+
+    This demo accompanies the paper ["Towards Better Characterization of Paraphrases" (ACL 2022)](https://github.com/tlkh/paraphrase-metrics), which describes in detail the methodologies used.""")
+
+with st.sidebar.expander("⚙️ Dataset Options", expanded=False):
+    st.markdown("This allows you to switch between the MRPC train and test sets, as well as choose to display only the original paraphrase pairs (MRPC) and/or the corrected pairs (MRPC-R1).")
     split = st.selectbox("Dataset Split", ["train", "test"])
-    display = st.selectbox("Source", ["All", "Only MRPC", "Only MRPC-R1"])
+    display = st.selectbox("Display only pairs from", ["All", "Only MRPC", "Only MRPC-R1"])
 
 ptype = st.sidebar.radio("Display Types", ["All",
                                            "Only Paraphrases (MRPC-R1)",
@@ -39,9 +43,9 @@ def load_df(split):
 
 def filter_df(df, display, ptype, filter_by, display_scores):
     # filter data
-    if display == "MRPC":
+    if display == "Only MRPC":
         df = df.drop(["new_s1", "new_s2"], axis=1)
-    elif display == "MRPC-R1":
+    elif display == "Only MRPC-R1":
         df = df.drop(["og_s1", "og_s2"], axis=1)
     # filter paraphrase type
     if ptype == "Only Paraphrases (MRPC)":