|
|
from CODE.PPLUIE.config import model_dict |
|
|
import gradio as gr |
|
|
|
|
|
def show_available_llms(): |
|
|
chaine = "" |
|
|
for k in model_dict.keys(): |
|
|
chaine+="\t"+ k + "\n" |
|
|
return f"Available models with ParaPLUIE: \n\n{chaine}" |
|
|
|
|
|
with gr.Blocks() as demo: |
|
|
gr.Markdown( |
|
|
""" |
|
|
# W.I.P |
|
|
""") |
|
|
gr.Markdown( |
|
|
""" |
|
|
# ParaPLUIE (Paraphrase Generation Evaluation Powered by an LLM) |
|
|
ParaPLUIE is a metric for evaluating the semantic proximity of two sentences. |
|
|
ParaPLUIE use the perplexity of an LLM to compute a confidence score. |
|
|
It has shown the highest correlation with human judgement on paraphrase classification meanwhile reamin the computional cost low as it roughtly equal to one token generation cost. |
|
|
""") |
|
|
text_box = gr.Textbox(show_available_llms(), show_label=False) |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
with gr.Row(): |
|
|
with gr.Column(scale=3): |
|
|
text1 = gr.Textbox(placeholder="Have you ever seen a tsunami ?", label="Source") |
|
|
text2 = gr.Textbox(placeholder="Have you ever seen a tiramisu ?", label="Hypothesis") |
|
|
with gr.Column(scale=1): |
|
|
btn1 = gr.Button("Compute") |
|
|
score = gr.Textbox(label="Score") |
|
|
demo.launch() |
|
|
|
