Update README.md

README.md

@@ -183,7 +183,7 @@ def make_reranker_input(t, q):
     return f"<<<Context>>>\n{t}\n\n<<<Query>>>\n{q}"
 
 def make_reranker_inference_conversation(context, question):
-    system_message = "Given a query and a piece of text, output a score of 1-7 based on how related the query is to the text. 1 means least related and 7 is most related."
+    system_message = "Given a piece of text and a query, output a score of 1-7 based on how related the query is to the text. 1 means least related and 7 is most related."
 
     return [
         {"role": "system", "content": system_message},
@@ -237,7 +237,7 @@ def make_reranker_input(t, q):
     return f"<<<Context>>>\n{t}\n\n<<<Query>>>\n{q}"
 
 def make_reranker_inference_conversation(context, question):
-    system_message = "Given a query and a piece of text, output a score of 1-7 based on how related the query is to the text. 1 means least related and 7 is most related."
+    system_message = "Given a piece of text and a query, output a score of 1-7 based on how related the query is to the text. 1 means least related and 7 is most related."
 
     return [
         {"role": "system", "content": system_message},
@@ -302,7 +302,7 @@ def make_reranker_input(t, q):
     return f"<<<Context>>>\n{t}\n\n<<<Query>>>\n{q}"
 
 def make_reranker_inference_conversation(context, question):
-    system_message = "Given a query and a piece of text, output a score of 1-7 based on how related the query is to the text. 1 means least related and 7 is most related."
+    system_message = "Given a piece of text and a query, output a score of 1-7 based on how related the query is to the text. 1 means least related and 7 is most related."
 
     return [
         {"role": "system", "content": system_message},
@@ -345,36 +345,6 @@ print(expected_vals)
 </details></li>
 </ul>	
 
-# Evaluation
-
-We perform an evaluation on 9 datasets from the [BEIR benchmark](https://github.com/beir-cellar/beir) that none of the evaluated models have been trained upon (to our knowledge).
-
-* Arguana
-* Dbpedia-entity
-* Fiqa
-* NFcorpus
-* Scidocs
-* Scifact
-* Trec-covid-v2
-* Vihealthqa
-* Webis-touche2020
-
-We evaluate on a subset of all queries (the first 250) to save evaluation time.
-
-We find that our model performs similarly or better than many of the state-of-the-art reranker models in our evaluation, without compromising on inference speed.
-
-We make our evaluation code and results available [on our Github](https://github.com/lightblue-tech/lb-reranker/blob/main/run_bier.ipynb).
-
-![image/png](https://cdn-uploads.huggingface.co/production/uploads/64b63f8ad57e02621dc93c8b/xkNzCABFUmU7UmDXUduiz.png)
-
-![image/png](https://cdn-uploads.huggingface.co/production/uploads/64b63f8ad57e02621dc93c8b/P-XCA3TGHqDSX8k6c4hCE.png)
-
-As we can see, this reranker attains greater IR evaluation metrics compared to the two benchmarks we include for all positions apart from @1.
-
-![image/png](https://cdn-uploads.huggingface.co/production/uploads/64b63f8ad57e02621dc93c8b/puhhWseBOcIyOEdW4L-B0.png)
-
-We also show that our model is, on average, faster than the BGE reranker v2.
-
 # License
 
 We share this model under an Apache 2.0 license.