Update main.py

main.py

@@ -42,7 +42,7 @@ def search(query: str, k: int) -> tuple[str]:
 
     search_results = "You are an AI assistant who delights in helping people" \
         + "learn about research from the Design Research Collective. Here are" \
-        + "several really cool abstracts:\n\n"
+        + "several abstracts from really cool, and really relevant, papers:\n\n"
 
     references = "\n\n## References\n\n"
 
@@ -51,13 +51,13 @@ def search(query: str, k: int) -> tuple[str]:
         references += str(i+1) + ". " + ", ".join([author.split(" ")[-1] for author in top_five["bib_dict"].values[i]["author"].split(" and ")]) + ". (" + str(int(top_five["bib_dict"].values[i]["pub_year"])) +  "). [" + top_five["bib_dict"].values[i]["title"] + "]" \
             + "(https://scholar.google.com/citations?view_op=view_citation&citation_for_view=" + top_five["author_pub_id"].values[i] + ").\n"
 
-    search_results += "\nSummarize the above abstracts as you respond to the following query:"
+    search_results += "\nResponse to the following query from the perspective of tehse examples only:"
 
     return search_results, references
 
 
 # Create an LLM pipeline that we can send queries to
-model_name = "Qwen/Qwen2-1.5B-Instruct"
+model_name = "Qwen/Qwen2-7B-Instruct"
 tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
 streamer = transformers.TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
 chatmodel = transformers.AutoModelForCausalLM.from_pretrained(