pages/18_DenseConnector.py

import streamlit as st
from streamlit_extras.switch_page_button import switch_page


translations = {
'en': {'title': 'DenseConnector',
    'original_tweet': 
       """
       [Original tweet](https://twitter.com/mervenoyann/status/1796089181988352216) (May 30, 2024)
       """,
    'tweet_1':
        """
        Do we fully leverage image encoders in vision language models? 👀  
        A new paper built a dense connector that does it better! Let's dig in 🧶 
        """,
    'tweet_2':
        """
        VLMs consist of an image encoder block, a projection layer that projects image embeddings to text embedding space and then a text decoder sequentially connected 📖  
        This [paper](https://t.co/DPQzbj0eWm) explores using intermediate states of image encoder and not a single output 🤩 
        """,
    'tweet_3':
        """
        The authors explore three different ways of instantiating dense connector: sparse token integration, sparse channel integration and dense channel integration (each of them just take intermediate outputs and put them together in different ways, see below).  
        """,
    'tweet_4':
        """
        They explore all three of them integrated to LLaVA 1.5 and found out each of the new models are superior to the original LLaVA 1.5.  
        """,
    'tweet_5':
        """
        I tried the [model](https://huggingface.co/spaces/HuanjinYao/DenseConnector-v1.5-8B) and it seems to work very well 🥹  
        The authors have released various [checkpoints](https://t.co/iF8zM2qvDa) based on different decoders (Vicuna 7/13B and Llama 3-8B). 
        """,
    'ressources':
        """
        Ressources:  
        [Dense Connector for MLLMs](https://arxiv.org/abs/2405.13800) 
        by Huanjin Yao, Wenhao Wu, Taojiannan Yang, YuXin Song, Mengxi Zhang, Haocheng Feng, Yifan Sun, Zhiheng Li, Wanli Ouyang, Jingdong Wang (2024)  
        [GitHub](https://github.com/HJYao00/DenseConnector)
        """
      },
'fr': {
    'title': 'DenseConnector',
    'original_tweet': 
       """
       [Tweet de base](https://twitter.com/mervenoyann/status/1796089181988352216) (en anglais) (30 mai 2024)
       """,
    'tweet_1':
        """
        Exploitons-nous pleinement les encodeurs d'images dans les modèles de langage/vision ? 👀  
        Un nouveau papier a construit un connecteur dense qui le fait mieux ! Creusons un peu 🧶 
        """,
    'tweet_2':
        """
        Les VLM se composent d'un bloc encodeur d'images, d'une couche de projection qui projette les enchâssements d'images dans l'espace d'enchâssement du texte, puis d'un décodeur de texte connecté séquentiellement 📖.  
        Ce [papier](https://t.co/DPQzbj0eWm) explore l'utilisation d'états intermédiaires de l'encodeur d'images et non d'une sortie unique 🤩 
        """,
    'tweet_3':
        """
        Les auteurs explorent trois manières différentes d'instancier un connecteur dense : l'intégration de tokens épars, l'intégration de canaux épars et l'intégration de canaux denses (chacune d'entre elles prend simplement des sorties intermédiaires et les rassemble de différentes manières, voir ci-dessous).  
        """,
    'tweet_4':
        """    
        Ils ont exploré les trois modèles intégrés à LLaVA 1.5 et ont constaté que chacun des nouveaux modèles est supérieur au LLaVA 1.5 original.  
        """,
    'tweet_5':
        """
        J'ai essayé le [modèle](https://huggingface.co/spaces/HuanjinYao/DenseConnector-v1.5-8B) et il semble fonctionner très bien 🥹  
        Les auteurs ont publié plusieurs [checkpoints](https://t.co/iF8zM2qvDa) basés sur différents décodeurs (Vicuna 7/13B et Llama 3-8B). 
        """,
    'ressources':
        """
        Ressources :  
        [Dense Connector for MLLMs](https://arxiv.org/abs/2405.13800) 
        de Huanjin Yao, Wenhao Wu, Taojiannan Yang, YuXin Song, Mengxi Zhang, Haocheng Feng, Yifan Sun, Zhiheng Li, Wanli Ouyang, Jingdong Wang (2024)  
        [GitHub](https://github.com/HJYao00/DenseConnector)
        """
    }
}    


def language_selector():
    languages = {'EN': '🇬🇧', 'FR': '🇫🇷'}
    selected_lang = st.selectbox('', options=list(languages.keys()), format_func=lambda x: languages[x], key='lang_selector')
    return 'en' if selected_lang == 'EN' else 'fr'

left_column, right_column = st.columns([5, 1])

# Add a selector to the right column
with right_column:
    lang = language_selector()

# Add a title to the left column
with left_column:
    st.title(translations[lang]["title"])
    
st.success(translations[lang]["original_tweet"], icon="ℹ️")
st.markdown(""" """)

st.markdown(translations[lang]["tweet_1"], unsafe_allow_html=True)
st.markdown(""" """)

st.image("pages/DenseConnector/image_1.jpg", use_column_width=True)
st.markdown(""" """)

st.markdown(translations[lang]["tweet_2"], unsafe_allow_html=True)
st.markdown(""" """)

st.image("pages/DenseConnector/image_2.jpg", use_column_width=True)
st.markdown(""" """)

st.markdown(translations[lang]["tweet_3"], unsafe_allow_html=True)
st.markdown(""" """)

st.image("pages/DenseConnector/image_3.jpg", use_column_width=True)
st.markdown(""" """)

st.markdown(translations[lang]["tweet_4"], unsafe_allow_html=True)
st.markdown(""" """)

st.image("pages/DenseConnector/image_4.jpg", use_column_width=True)
st.markdown(""" """)

st.markdown(translations[lang]["tweet_5"], unsafe_allow_html=True)
st.markdown(""" """)

st.image("pages/DenseConnector/image_5.jpg", use_column_width=True)
st.markdown(""" """)

st.info(translations[lang]["ressources"], icon="📚")  

st.markdown(""" """)
st.markdown(""" """)
st.markdown(""" """)
col1, col2, col3= st.columns(3)
with col1:
    if lang == "en":
        if st.button('Previous paper', use_container_width=True):
            switch_page("CuMo")
    else:
        if st.button('Papier précédent', use_container_width=True):
            switch_page("CuMo")
with col2:
    if lang == "en":
        if st.button("Home", use_container_width=True):
            switch_page("Home")
    else:
        if st.button("Accueil", use_container_width=True):
            switch_page("Home")
with col3:
    if lang == "en":
        if st.button("Next paper", use_container_width=True):
            switch_page("Depth Anything v2")
    else:
        if st.button("Papier suivant", use_container_width=True):
            switch_page("Depth Anything v2")