Event: llm_response

Description: Response from our language model that has been generated so far.

Example:

{
  event: "llm_response",
  id: "your_conversation_id",
  text: "Yes, cancer can be transmissible however, ..."
}

Event: articles

Description: Articles that are referenced in the response of MediSearch. Arrives after the llm_response is done generating.

Example:

{
	event: "articles",
	id: "your_conversation_id",
  articles: [
   {
      "title":"A review of the evaluation and management of cartilage defects in the knee.",
      "url":"https://pubmed.ncbi.nlm.nih.gov/21378492/",
      "authors":[
         "Henn",
         "Gomoll"
      ],
      "year":"2013",
   },
	]
}

The citations in the llm_response correspond to the 1-indexed articles in the returned articles list. For example, if the response from articles was

articles_response = {...

  articles: [
   {...
      "title":"Cancer is not transmissible",
   },
	 {
      "title":"Transmissible cancer can actually happen",
   },
	]
}

and the response from llm_response was

{...
  text: "No, cancer is not transmissible [1]. Except for some rare cases [2]."
}

then the [1] corresponds to “Cancer is not transmissible” and the [2] corresponds to "Transmissible cancer can actually happen" . In other words [1] corresponds to articles_response.articles[0] and [2] corresponds to articles_response.articles[1]. In general: the i-th citation in the response coming from llm_response corresponds to the (i-1)st index of the articles list coming from articles. Phew—sorry for this, but not sure if there’s an easier way.