Spaces:

AnishaBhatnagar
/

LLVM_trial

Paused

Anisha Bhatnagar commited on Dec 11, 2023

Commit

125208c

1 Parent(s): daeb1c8

removed image

Files changed (1) hide show

app.py CHANGED Viewed

@@ -110,8 +110,8 @@ print ('Step2: feed-forward process')
 #title = "Hate-LLaMA - An Instruction-tuned Audio-Visual Language Model for Hate Content Detection"
 description = """
-<h1 align="center"> Hate-LLaMA <img src="/file=hate_llama_icon.png", border="0" style="margin: 0 auto; height: 200px;" /></h1>
-<h4 align="center">  An Audio-Visual Language Model for Hate Content Detection </h4>
 Hate-LLaMA , is a multi-modal framework, designed to detect hate in videos and classify them as HATE or NON HATE. Hate-LLaMA finetunes Video-LLaMA (which uses the LLaMA-7b-chat model as backbone). The model is able to analyse both the audio and visual content to perform the classification task. After uploading a video and clicking submit, the model outputs a simple statement identifying if the video has hate or not.

 #title = "Hate-LLaMA - An Instruction-tuned Audio-Visual Language Model for Hate Content Detection"
 description = """
+<h1 align="center"> Hate-LLaMA </h1>
+<h3 align="center">  An Audio-Visual Language Model for Hate Content Detection </h3>
 Hate-LLaMA , is a multi-modal framework, designed to detect hate in videos and classify them as HATE or NON HATE. Hate-LLaMA finetunes Video-LLaMA (which uses the LLaMA-7b-chat model as backbone). The model is able to analyse both the audio and visual content to perform the classification task. After uploading a video and clicking submit, the model outputs a simple statement identifying if the video has hate or not.