Update README.md

README.md

@@ -3,28 +3,23 @@ license: cc-by-sa-4.0
 inference: false
 ---
 
-# SLIM-SA-NER-3B
+# SLIM-EXTRACT
 
 <!-- Provide a quick summary of what the model is/does. -->
 
-**slim-sa-ner-3b** combines two of the most popular traditional classifier functions (**Sentiment Analysis** and **Named Entity Recognition**), and reimagines them as function calls on a specialized decoder-based LLM, generating output consisting of a python dictionary with keys corresponding to sentiment, and NER identifiers, such as people, organization, and place, e.g.:  
-
-&nbsp;&nbsp;&nbsp;&nbsp;`{'sentiment': ['positive'], people': ['..'], 'organization': ['..'],'place': ['..]}`
-
-This 'combo' model is designed to illustrate the potential power of using function calls on small, specialized models to enable a single model architecture to combine the capabilities of what were traditionally two separate model architectures on an encoder.  
-
-The intent of SLIMs is to forge a middle-ground between traditional encoder-based classifiers and open-ended API-based LLMs, providing an intuitive, flexible natural language response, without complex prompting, and with improved generalization and ability to fine-tune to a specific domain use case.  
+**slim-extract** implements a specialized function-calling customizable 'extract' capability that takes as an input a context passage, a customized key, and outputs a python dictionary with key that corresponds to the customized key, with a value consisting of a list of items extracted from the text corresponding to that key, e.g., 
 
+&nbsp;&nbsp;&nbsp;&nbsp;`{'universities': ['Berkeley, Stanford, Yale, University of Florida, ...'] }`
 
 This model is fine-tuned on top of [**llmware/bling-stable-lm-3b-4e1t-v0**](https://huggingface.co/llmware/bling-stable-lm-3b-4e1t-v0), which in turn, is a fine-tune of stabilityai/stablelm-3b-4elt.
 
-Each slim model has a 'quantized tool' version, e.g.,  [**'slim-sa-ner-3b-tool'**](https://huggingface.co/llmware/slim-sa-ner-3b-tool).  
+For fast inference use, we would recommend the 'quantized tool' version, e.g.,  [**'slim-extract-tool'**](https://huggingface.co/llmware/slim-sa-ner-3b-tool).  
 
 
 ## Prompt format:
 
-`function = "classify"`  
-`params = "sentiment, person, organization, place"`  
+`function = "extract"`  
+`params = "{custom key}"`  
 `prompt = "<human> " + {text} + "\n" + `  
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp; &nbsp;`"<{function}> " + {params} + "</{function}>" + "\n<bot>:"`  
 
@@ -32,11 +27,11 @@ Each slim model has a 'quantized tool' version, e.g.,  [**'slim-sa-ner-3b-tool'*
 <details>
 <summary>Transformers Script </summary>
 
-    model = AutoModelForCausalLM.from_pretrained("llmware/slim-sa-ner-3b")
-    tokenizer = AutoTokenizer.from_pretrained("llmware/slim-sa-ner-3b")
+    model = AutoModelForCausalLM.from_pretrained("llmware/slim-extract")
+    tokenizer = AutoTokenizer.from_pretrained("llmware/slim-extract")
 
-    function = "classify"
-    params = "topic"
+    function = "extract"
+    params = "company"
 
     text = "Tesla stock declined yesterday 8% in premarket trading after a poorly-received event in San Francisco yesterday, in which the company indicated a likely shortfall in revenue."  
     
@@ -75,8 +70,8 @@ Each slim model has a 'quantized tool' version, e.g.,  [**'slim-sa-ner-3b-tool'*
 <summary>Using as Function Call in LLMWare</summary>
 
     from llmware.models import ModelCatalog
-    slim_model = ModelCatalog().load_model("llmware/slim-sa-ner-3b")
-    response = slim_model.function_call(text,params=["sentiment", "people", "organization", "place"], function="classify")
+    slim_model = ModelCatalog().load_model("llmware/slim-extract")
+    response = slim_model.function_call(text,params=["company"], function="extract")
 
     print("llmware - llm_response: ", response)