Update app.py

app.py

@@ -210,24 +210,15 @@ demo = gr.Blocks()
 with demo:
     input_texts = gr.Variable([])
     gr.Markdown("**Detect Task Specification at Inference-time.**")
-    # gr.Markdown("LLMs are pretty good at reporting task underspecification. We just need to ask the right way.")
-    # gr.Markdown("Using our Underspecification Metric informed by applying causal inference techniques, \
-    #     we are able to identify likely spurious correlations and exploit them in \
-    #     the scenario of gender underspecified tasks. (Note that introspecting softmax probabilities alone is insufficient, as in the sentences \
-    #     below, LLMs may report a softmax prob of ~0.9 despite the task being underspecified.)")
-    # gr.Markdown("We extend the [Winogender Schemas](https://github.com/rudinger/winogender-schemas) evaluation set to produce\
-    #     eight syntactically similar sentences. However semantically, \
-    #     only two of the sentences are well-specified while the rest remain underspecified.")
-    # gr.Markdown("If a model can reliably report the underspecification of an inference-time task, an AI systems can replace only those task predictions with\
-    #     an appropriate heuristic or information retrieval process.")
-    gr.Markdown("*Follow the numbered steps below to test one of the pre-loaded options.* Once you get the hang of it, you can load a new model and/or provide your own input texts.")
-    gr.Markdown(f"""1) Pick a preloaded BERT-like model. 
-        Note: RoBERTa-large performance is best.
-    2) Pick an Occupation type from the Winogender Schemas evaluation set.
-        Or select '{PICK_YOUR_OWN_LABEL}' (it need not be about an occupation).
-    3) Click button to load input texts.
-        Read the sentences to determine which two are well-specified for gendered pronoun coreference resolution. The rest are gender-unspecified.
-    4) Click button to get Task Specification Metric results!
+
+    gr.Markdown("**Follow the numbered steps below to test one of the pre-loaded options.** Once you get the hang of it, you can load a new model and/or provide your own input texts.")
+    gr.Markdown(f"""1) Pick a preloaded BERT-like model.  
+        *Note: RoBERTa-large performance is best.*
+    2) Pick an Occupation type from the Winogender Schemas evaluation set.   
+        *Or select '{PICK_YOUR_OWN_LABEL}' (it need not be about an occupation).*
+    3) Click button to load input texts.   
+        *Read the sentences to determine which two are well-specified for gendered pronoun coreference resolution. The rest are gender-unspecified.*
+    4) Click button to get Task Specification Metric results!   
     """)
     
 
@@ -258,7 +249,7 @@ with demo:
         )
 
     with gr.Row():
-        get_text_btn = gr.Button("3) Click to load input texts.\n(Read the sentences to determine which two are well-specified for gendered pronoun coreference resolution. The rest are gender-unspecified.)")
+        get_text_btn = gr.Button("3) Click to load input texts.)")
 
     get_text_btn.click(
         fn=display_input_texts,