put config at bottom of readme

Leaderboard.md

@@ -1,163 +0,0 @@
-# 🔥🏅️GenCeption Leaderboard 🏅️🔥
-
-Evaluated MLLMs: [ChatGPT-4V](https://cdn.openai.com/papers/GPTV_System_Card.pdf), [mPLUG-Owl2](https://arxiv.org/pdf/2311.04257.pdf), [LLaVA-13B](https://arxiv.org/pdf/2304.08485.pdf), [LLaVA-7B](https://arxiv.org/pdf/2304.08485.pdf)
-
-<table>
-<tr><th>Existence </th><th>Count</th></tr>
-<tr><td>
-
-| Model | GC@3|
-|--|--|
-| ChatGPT-4V|0.422 |
-| mPLUG-Owl2|0.323 |
-| LLaVA-7B|0.308 |
-| LLaVA-13B|0.305 |
-
-</td><td>
-
-| Model | GC@3|
-|--|--|
-| ChatGPT-4V|0.404 |
-| mPLUG-Owl2|0.299 |
-| LLaVA-13B|0.294 |
-| LLaVA-7B|0.353 |
-
-</td></tr> </table>
-
-
-<table>
-<tr><th>Position </th><th>Color</th></tr>
-<tr><td>
-
-| Model | GC@3|
-|--|--|
-| ChatGPT-4V|0.408|
-| mPLUG-Owl2|0.306 |
-| LLaVA-7B|0.285 |
-| LLaVA-13B|0.255 |
-
-</td><td>
-
-| Model | GC@3|
-|--|--|
-| ChatGPT-4V|0.403 |
-| LLaVA-13B|0.300 |
-| mPLUG-Owl2|0.290 |
-| LLaVA-7B|0.284 |
-
-</td></tr> </table>
-
-
-<table>
-<tr><th>Poster </th><th>Celebrity</th></tr>
-<tr><td>
-
-| Model | GC@3|
-|--|--|
-| ChatGPT-4V|0.324|
-| mPLUG-Owl2|0.243 |
-| LLaVA-13B|0.215 |
-| LLaVA-7B|0.214 |
-
-</td><td>
-
-| Model | GC@3|
-|--|--|
-| ChatGPT-4V|0.332 |
-| mPLUG-Owl2|0.232 |
-| LLaVA-13B|0.206 |
-| LLaVA-7B|0.188 |
-
-</td></tr> </table>
-
-
-<table>
-<tr><th>Scene </th><th>Landmark</th></tr>
-<tr><td>
-
-| Model | GC@3|
-|--|--|
-| ChatGPT-4V|0.393|
-| mPLUG-Owl2|0.299 |
-| LLaVA-13B|0.277 |
-| LLaVA-7B|0.266 |
-
-</td><td>
-
-| Model | GC@3|
-|--|--|
-| ChatGPT-4V|0.353 |
-| mPLUG-Owl2|0.275 |
-| LLaVA-7B|0.252 |
-| LLaVA-13B|0.242 |
-
-</td></tr> </table>
-
-
-<table>
-<tr><th>Artwork </th><th>Commonsense Reasoning</th></tr>
-<tr><td>
-
-| Model | GC@3|
-|--|--|
-| ChatGPT-4V|0.421|
-| mPLUG-Owl2|0.252 |
-| LLaVA-13B|0.212 |
-| LLaVA-7B|0.210 |
-
-</td><td>
-
-| Model | GC@3|
-|--|--|
-| ChatGPT-4V|0.471 |
-| mPLUG-Owl2|0.353 |
-| LLaVA-13B|0.334 |
-| LLaVA-7B|0.294 |
-
-</td></tr> </table>
-
-
-<table>
-<tr><th>Code Reasoning </th><th>Numerical Calculation</th></tr>
-<tr><td>
-
-| Model | GC@3|
-|--|--|
-| ChatGPT-4V|0.193|
-| mPLUG-Owl2|0.176 |
-| LLaVA-13B|0.144 |
-| LLaVA-7B|0.107 |
-
-</td><td>
-
-| Model | GC@3|
-|--|--|
-| ChatGPT-4V|0.240 |
-| LLaVA-13B|0.195 |
-| mPLUG-Owl2|0.192 |
-| LLaVA-7B|0.155 |
-
-</td></tr> </table>
-
-
-<table>
-<tr><th>Text Translation </th><th>OCR</th></tr>
-<tr><td>
-
-| Model | GC@3|
-|--|--|
-| ChatGPT-4V|0.157|
-| LLaVA-13B|0.116 |
-| LLaVA-7B|0.111 |
-| mPLUG-Owl2|0.081 |
-
-</td><td>
-
-| Model | GC@3|
-|--|--|
-| ChatGPT-4V|0.393 |
-| mPLUG-Owl2|0.276 |
-| LLaVA-13B|0.239 |
-| LLaVA-7B|0.222 |
-
-</td></tr> </table>

README.md

@@ -1,14 +1,3 @@
----
-title: Genception Leaderboard
-emoji: 🔥
-colorFrom: red
-colorTo: green
-sdk: gradio
-sdk_version: 4.19.2
-app_file: app.py
-pinned: true
----
-
 # GenCeption: Evaluate Multimodal LLMs with Unlabeled Unimodal Data
 
 <div>
@@ -33,7 +22,7 @@ We demostrate a 5-iteration GenCeption procedure below run on a seed images to e
 
 
 ## Contribute
-Please add your model details and results to `leaderboard/leaderboard.json` and **create a PR (Pull-Request)** to contribute your results to the [🔥🏅️**Leaderboard**🏅️🔥](https://huggingface.co/spaces/). Start by creating your virtual environment:
+Please add your model details and results to `leaderboard/leaderboard.json` and **create a PR (Pull-Request)** to contribute your results to the [🔥🏅️**Leaderboard**🏅️🔥](https://huggingface.co/spaces/valbuc/GenCeption). Start by creating your virtual environment:
 
 ```{bash}
 conda create --name genception python=3.10 -y
@@ -64,3 +53,18 @@ The MME dataset, of which the image modality was used in our paper, can be obtai
     primaryClass={cs.AI,cs.CL,cs.LG}
 }
 ```
+
+## HF Space config
+
+Please dont be distracted by this content - it just configues the [🤗 Leaderboard](https://huggingface.co/spaces/valbuc/GenCeption).
+
+---
+title: Genception Leaderboard
+emoji: 🔥
+colorFrom: red
+colorTo: green
+sdk: gradio
+sdk_version: 4.19.2
+app_file: app.py
+pinned: true
+---

app.py

@@ -104,4 +104,4 @@ scheduler = BackgroundScheduler()
 scheduler.add_job(update_data, "cron", hour=0)  # Update data once a day at midnight
 scheduler.start()
 
-demo.queue(default_concurrency_limit=40).launch(share=True)
+demo.queue(default_concurrency_limit=40).launch()