Refactor gradio Tabs initialization

.gitignore

@@ -0,0 +1 @@
+*.pyc

app.py

@@ -23,7 +23,7 @@ TASKS = [
 ]
 
 TASK_LIST_BITEXT_MINING = ['BUCC (de-en)', 'BUCC (fr-en)', 'BUCC (ru-en)', 'BUCC (zh-en)', 'Tatoeba (afr-eng)', 'Tatoeba (amh-eng)', 'Tatoeba (ang-eng)', 'Tatoeba (ara-eng)', 'Tatoeba (arq-eng)', 'Tatoeba (arz-eng)', 'Tatoeba (ast-eng)', 'Tatoeba (awa-eng)', 'Tatoeba (aze-eng)', 'Tatoeba (bel-eng)', 'Tatoeba (ben-eng)', 'Tatoeba (ber-eng)', 'Tatoeba (bos-eng)', 'Tatoeba (bre-eng)', 'Tatoeba (bul-eng)', 'Tatoeba (cat-eng)', 'Tatoeba (cbk-eng)', 'Tatoeba (ceb-eng)', 'Tatoeba (ces-eng)', 'Tatoeba (cha-eng)', 'Tatoeba (cmn-eng)', 'Tatoeba (cor-eng)', 'Tatoeba (csb-eng)', 'Tatoeba (cym-eng)', 'Tatoeba (dan-eng)', 'Tatoeba (deu-eng)', 'Tatoeba (dsb-eng)', 'Tatoeba (dtp-eng)', 'Tatoeba (ell-eng)', 'Tatoeba (epo-eng)', 'Tatoeba (est-eng)', 'Tatoeba (eus-eng)', 'Tatoeba (fao-eng)', 'Tatoeba (fin-eng)', 'Tatoeba (fra-eng)', 'Tatoeba (fry-eng)', 'Tatoeba (gla-eng)', 'Tatoeba (gle-eng)', 'Tatoeba (glg-eng)', 'Tatoeba (gsw-eng)', 'Tatoeba (heb-eng)', 'Tatoeba (hin-eng)', 'Tatoeba (hrv-eng)', 'Tatoeba (hsb-eng)', 'Tatoeba (hun-eng)', 'Tatoeba (hye-eng)', 'Tatoeba (ido-eng)', 'Tatoeba (ile-eng)', 'Tatoeba (ina-eng)', 'Tatoeba (ind-eng)', 'Tatoeba (isl-eng)', 'Tatoeba (ita-eng)', 'Tatoeba (jav-eng)', 'Tatoeba (jpn-eng)', 'Tatoeba (kab-eng)', 'Tatoeba (kat-eng)', 'Tatoeba (kaz-eng)', 'Tatoeba (khm-eng)', 'Tatoeba (kor-eng)', 'Tatoeba (kur-eng)', 'Tatoeba (kzj-eng)', 'Tatoeba (lat-eng)', 'Tatoeba (lfn-eng)', 'Tatoeba (lit-eng)', 'Tatoeba (lvs-eng)', 'Tatoeba (mal-eng)', 'Tatoeba (mar-eng)', 'Tatoeba (max-eng)', 'Tatoeba (mhr-eng)', 'Tatoeba (mkd-eng)', 'Tatoeba (mon-eng)', 'Tatoeba (nds-eng)', 'Tatoeba (nld-eng)', 'Tatoeba (nno-eng)', 'Tatoeba (nob-eng)', 'Tatoeba (nov-eng)', 'Tatoeba (oci-eng)', 'Tatoeba (orv-eng)', 'Tatoeba (pam-eng)', 'Tatoeba (pes-eng)', 'Tatoeba (pms-eng)', 'Tatoeba (pol-eng)', 'Tatoeba (por-eng)', 'Tatoeba (ron-eng)', 'Tatoeba (rus-eng)', 'Tatoeba (slk-eng)', 'Tatoeba (slv-eng)', 'Tatoeba (spa-eng)', 'Tatoeba (sqi-eng)', 'Tatoeba (srp-eng)', 'Tatoeba (swe-eng)', 'Tatoeba (swg-eng)', 'Tatoeba (swh-eng)', 'Tatoeba (tam-eng)', 'Tatoeba (tat-eng)', 'Tatoeba (tel-eng)', 'Tatoeba (tgl-eng)', 'Tatoeba (tha-eng)', 'Tatoeba (tuk-eng)', 'Tatoeba (tur-eng)', 'Tatoeba (tzl-eng)', 'Tatoeba (uig-eng)', 'Tatoeba (ukr-eng)', 'Tatoeba (urd-eng)', 'Tatoeba (uzb-eng)', 'Tatoeba (vie-eng)', 'Tatoeba (war-eng)', 'Tatoeba (wuu-eng)', 'Tatoeba (xho-eng)', 'Tatoeba (yid-eng)', 'Tatoeba (yue-eng)', 'Tatoeba (zsm-eng)']
-TASK_LIST_BITEXT_MINING_OTHER = ["BornholmBitextMining"]
+TASK_LIST_BITEXT_MINING_DA = ["BornholmBitextMining"]
 
 TASK_LIST_CLASSIFICATION = [
     "AmazonCounterfactualClassification (en)",
@@ -1027,7 +1027,7 @@ def add_task(examples):
         examples["mteb_task"] = "STS"
     elif examples["mteb_dataset_name"] in norm(TASK_LIST_SUMMARIZATION + TASK_LIST_SUMMARIZATION_FR):
         examples["mteb_task"] = "Summarization"
-    elif examples["mteb_dataset_name"] in norm(TASK_LIST_BITEXT_MINING + TASK_LIST_BITEXT_MINING_OTHER):
+    elif examples["mteb_dataset_name"] in norm(TASK_LIST_BITEXT_MINING + TASK_LIST_BITEXT_MINING_DA):
         examples["mteb_task"] = "BitextMining"
     else:
         print("WARNING: Task not found for dataset", examples["mteb_dataset_name"])
@@ -1427,7 +1427,7 @@ get_mteb_average_fr()
 get_mteb_average_pl()
 get_mteb_average_zh()
 DATA_BITEXT_MINING = get_mteb_data(["BitextMining"], [], TASK_LIST_BITEXT_MINING)
-DATA_BITEXT_MINING_OTHER = get_mteb_data(["BitextMining"], [], TASK_LIST_BITEXT_MINING_OTHER)
+DATA_BITEXT_MINING_DA = get_mteb_data(["BitextMining"], [], TASK_LIST_BITEXT_MINING_DA)
 DATA_CLASSIFICATION_DA = get_mteb_data(["Classification"], [], TASK_LIST_CLASSIFICATION_DA)
 DATA_CLASSIFICATION_NB = get_mteb_data(["Classification"], [], TASK_LIST_CLASSIFICATION_NB)
 DATA_CLASSIFICATION_SV = get_mteb_data(["Classification"], [], TASK_LIST_CLASSIFICATION_SV)
@@ -1442,7 +1442,7 @@ MODELS = []
 # LANGUAGES = []
 for d in [
     DATA_BITEXT_MINING,
-    DATA_BITEXT_MINING_OTHER,
+    DATA_BITEXT_MINING_DA,
     DATA_CLASSIFICATION_EN,
     DATA_CLASSIFICATION_DA,
     DATA_CLASSIFICATION_FR,
@@ -1505,783 +1505,346 @@ table > tbody > tr > td:nth-child(2) > div {
 }
 """
 
-block = gr.Blocks(css=css)
-with block:
-    gr.Markdown(f"""
-    Massive Text Embedding Benchmark (MTEB) Leaderboard. To submit, refer to the <a href="https://github.com/embeddings-benchmark/mteb#leaderboard" target="_blank" style="text-decoration: underline">MTEB GitHub repository</a> 🤗 Refer to the [MTEB paper](https://arxiv.org/abs/2210.07316) for details on metrics, tasks and models.
-    """)
+"""
+Each inner tab can have the following keys:
+- language: The language of the leaderboard
+- language_long: [optional] The long form of the language
+- description: The description of the leaderboard
+- credits: [optional] The credits for the leaderboard
+- data: The data for the leaderboard
+- refresh: The function to refresh the leaderboard
+"""
+
+chinese_credits = "[FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding)"
+french_credits = "[Lyon-NLP](https://github.com/Lyon-NLP): [Gabriel Sequeira](https://github.com/GabrielSequeira), [Imene Kerboua](https://github.com/imenelydiaker), [Wissam Siblini](https://github.com/wissam-sib), [Mathieu Ciancone](https://github.com/MathieuCiancone), [Marion Schaeffer](https://github.com/schmarion)"
+danish_credits = "[Kenneth Enevoldsen](https://github.com/KennethEnevoldsen), [scandinavian-embedding-benchmark](https://kennethenevoldsen.github.io/scandinavian-embedding-benchmark/)"
+norwegian_credits = "[Kenneth Enevoldsen](https://github.com/KennethEnevoldsen), [scandinavian-embedding-benchmark](https://kennethenevoldsen.github.io/scandinavian-embedding-benchmark/)"
+polish_credits = "[Rafał Poświata](https://github.com/rafalposwiata)"
+
+data = {
+    "Overall": {
+        "metric": "Various, refer to task tabs",
+        "data": [
+            {
+                "language": "English",
+                "description": "**Overall MTEB English leaderboard** 🔮",
+                "data": DATA_OVERALL,
+                "refresh": get_mteb_average,
+            },
+            {
+                "language": "Chinese",
+                "data": DATA_OVERALL_ZH,
+                "description": "**Overall MTEB Chinese leaderboard (C-MTEB)** 🔮🇨🇳",
+                "credits": chinese_credits,
+                "refresh": get_mteb_average_zh,
+            },
+            {
+                "language": "French",
+                "data": DATA_OVERALL_FR,
+                "description": "**Overall MTEB French leaderboard (F-MTEB)** 🔮🇫🇷",
+                "credits": french_credits,
+                "refresh": get_mteb_average_fr,
+            },
+            {
+                "language": "Polish",
+                "data": DATA_OVERALL_PL,
+                "description": "**Overall MTEB Polish leaderboard** 🔮🇵🇱",
+                "refresh": get_mteb_average_pl,
+            },
+        ]
+    },
+    "Bitext Mining": {
+        "metric": "[F1](https://huggingface.co/spaces/evaluate-metric/f1)",
+        "data": [
+            {
+                "language": "English-X",
+                "language_long": "117 (Pairs of: English & other language)",
+                "description": "**Bitext Mining English-X Leaderboard** 🎌",
+                "data": DATA_BITEXT_MINING,
+                "refresh": partial(get_mteb_data, tasks=["BitextMining"], datasets=TASK_LIST_BITEXT_MINING),
+            },
+            {
+                "language": "Danish",
+                "language_long": "Danish & Bornholmsk (Danish Dialect)",
+                "description": "**Bitext Mining Danish Leaderboard** 🎌🇩🇰",
+                "credits": danish_credits,
+                "data": DATA_BITEXT_MINING_DA,
+                "refresh": partial(get_mteb_data, tasks=["BitextMining"], datasets=TASK_LIST_BITEXT_MINING_DA),
+            }
+        ]
+    },
+    "Classification": {
+        "metric": "[Accuracy](https://huggingface.co/spaces/evaluate-metric/accuracy)",
+        "data": [
+            {
+                "language": "English",
+                "description": "**Classification English Leaderboard** ❤️",
+                "data": DATA_CLASSIFICATION_EN,
+                "refresh": partial(get_mteb_data, tasks=["Classification"], langs=["en"])
+            },
+            {
+                "language": "Chinese",
+                "description": "**Classification Chinese Leaderboard** 🧡🇨🇳",
+                "credits": chinese_credits,
+                "data": DATA_CLASSIFICATION_ZH,
+                "refresh": partial(get_mteb_data, tasks=["Classification"], datasets=TASK_LIST_CLASSIFICATION_ZH)
+            },
+            {
+                "language": "Danish",
+                "description": "**Classification Danish Leaderboard** 🤍🇩🇰",
+                "credits": danish_credits,
+                "data": DATA_CLASSIFICATION_DA,
+                "refresh": partial(get_mteb_data, tasks=["Classification"], datasets=TASK_LIST_CLASSIFICATION_DA)
+            },
+            {
+                "language": "French",
+                "description": "**Classification French Leaderboard** 💙🇫🇷",
+                "credits": french_credits,
+                "data": DATA_CLASSIFICATION_FR,
+                "refresh": partial(get_mteb_data, tasks=["Classification"], datasets=TASK_LIST_CLASSIFICATION_FR)
+            },
+            {
+                "language": "Norwegian",
+                "language_long": "Norwegian Bokmål",
+                "description": "**Classification Norwegian Leaderboard** 💙🇳🇴",
+                "credits": norwegian_credits,
+                "data": DATA_CLASSIFICATION_NB,
+                "refresh": partial(get_mteb_data, tasks=["Classification"], datasets=TASK_LIST_CLASSIFICATION_NB)
+            },
+            {
+                "language": "Polish",
+                "description": "**Classification Polish Leaderboard** 🤍🇵🇱",
+                "credits": polish_credits,
+                "data": DATA_CLASSIFICATION_PL,
+                "refresh": partial(get_mteb_data, tasks=["Classification"], datasets=TASK_LIST_CLASSIFICATION_PL)
+            },
+            {
+                "language": "Swedish",
+                "description": "**Classification Swedish Leaderboard** 💛🇸🇪",
+                "credits": norwegian_credits,
+                "data": DATA_CLASSIFICATION_SV,
+                "refresh": partial(get_mteb_data, tasks=["Classification"], datasets=TASK_LIST_CLASSIFICATION_SV)
+            },
+            {
+                "language": "Other",
+                "language_long": "47 (Only languages not included in the other tabs)",
+                "description": "**Classification Other Languages Leaderboard** 💜💚💙",
+                "data": DATA_CLASSIFICATION_OTHER,
+                "refresh": partial(get_mteb_data, tasks=["Classification"], datasets=TASK_LIST_CLASSIFICATION_OTHER)
+            }
+        ]
+    },
+    "Clustering": {
+        "metric": "Validity Measure (v_measure)",
+        "data": [
+            {
+                "language": "English",
+                "description": "**Clustering Leaderboard** ✨",
+                "data": DATA_CLUSTERING,
+                "refresh": partial(get_mteb_data, tasks=["Clustering"], datasets=TASK_LIST_CLUSTERING)
+            },
+            {
+                "language": "Chinese",
+                "description": "**Clustering Chinese Leaderboard** ✨🇨🇳",
+                "credits": chinese_credits,
+                "data": DATA_CLUSTERING_ZH,
+                "refresh": partial(get_mteb_data, tasks=["Clustering"], datasets=TASK_LIST_CLUSTERING_ZH)
+            },
+            {
+                "language": "French",
+                "description": "**Clustering French Leaderboard** ✨🇫🇷",
+                "credits": french_credits,
+                "data": DATA_CLUSTERING_FR,
+                "refresh": partial(get_mteb_data, tasks=["Clustering"], datasets=TASK_LIST_CLUSTERING_FR)
+            },
+            {
+                "language": "German",
+                "description": "**Clustering German Leaderboard** ✨🇩🇪",
+                "credits": "[Silvan](https://github.com/slvnwhrl)",
+                "data": DATA_CLUSTERING_DE,
+                "refresh": partial(get_mteb_data, tasks=["Clustering"], datasets=TASK_LIST_CLUSTERING_DE)
+            },
+            {
+                "language": "Polish",
+                "description": "**Clustering Polish Leaderboard** ✨🇵🇱",
+                "credits": polish_credits,
+                "data": DATA_CLUSTERING_PL,
+                "refresh": partial(get_mteb_data, tasks=["Clustering"], datasets=TASK_LIST_CLUSTERING_PL)
+            },
+        ]
+    },
+    "Pair Classification": {
+        "metric": "Average Precision based on Cosine Similarities (cos_sim_ap)",
+        "data": [
+            {
+                "language": "English",
+                "description": "**Pair Classification English Leaderboard** 🎭",
+                "data": DATA_PAIR_CLASSIFICATION,
+                "refresh": partial(get_mteb_data, tasks=["PairClassification"], datasets=TASK_LIST_PAIR_CLASSIFICATION)
+            },
+            {
+                "language": "Chinese",
+                "description": "**Pair Classification Chinese Leaderboard** 🎭🇨🇳",
+                "credits": chinese_credits,
+                "data": DATA_PAIR_CLASSIFICATION_ZH,
+                "refresh": partial(get_mteb_data, tasks=["PairClassification"], datasets=TASK_LIST_PAIR_CLASSIFICATION_ZH)
+            },
+            {
+                "language": "French",
+                "description": "**Pair Classification French Leaderboard** 🎭🇫🇷",
+                "credits": french_credits,
+                "data": DATA_PAIR_CLASSIFICATION_FR,
+                "refresh": partial(get_mteb_data, tasks=["PairClassification"], datasets=TASK_LIST_PAIR_CLASSIFICATION_FR)
+            },
+            {
+                "language": "Polish",
+                "description": "**Pair Classification Polish Leaderboard** 🎭🇵🇱",
+                "credits": polish_credits,
+                "data": DATA_PAIR_CLASSIFICATION_PL,
+                "refresh": partial(get_mteb_data, tasks=["PairClassification"], datasets=TASK_LIST_PAIR_CLASSIFICATION_PL)
+            },
+        ]
+    },
+    "Reranking": {
+        "metric": "Mean Average Precision (MAP)",
+        "data": [
+            {
+                "language": "English",
+                "description": "**Reranking English Leaderboard** 🥈",
+                "data": DATA_RERANKING,
+                "refresh": partial(get_mteb_data, tasks=["Reranking"], datasets=TASK_LIST_RERANKING)
+            },
+            {
+                "language": "Chinese",
+                "description": "**Reranking Chinese Leaderboard** 🥈🇨🇳",
+                "credits": chinese_credits,
+                "data": DATA_RERANKING_ZH,
+                "refresh": partial(get_mteb_data, tasks=["Reranking"], datasets=TASK_LIST_RERANKING_ZH)
+            },
+            {
+                "language": "French",
+                "description": "**Reranking French Leaderboard** 🥈🇫🇷",
+                "credits": french_credits,
+                "data": DATA_RERANKING_FR,
+                "refresh": partial(get_mteb_data, tasks=["Reranking"], datasets=TASK_LIST_RERANKING_FR)
+            }
+        ]
+    },
+    "Retrieval": {
+        "metric": "Normalized Discounted Cumulative Gain @ k (ndcg_at_10)",
+        "data": [
+            {
+                "language": "English",
+                "description": "**Retrieval English Leaderboard** 🔎",
+                "data": DATA_RETRIEVAL,
+                "refresh": partial(get_mteb_data, tasks=["Retrieval"], datasets=TASK_LIST_RETRIEVAL)
+            },
+            {
+                "language": "Chinese",
+                "description": "**Retrieval Chinese Leaderboard** 🔎🇨🇳",
+                "credits": chinese_credits,
+                "data": DATA_RETRIEVAL_ZH,
+                "refresh": partial(get_mteb_data, tasks=["Retrieval"], datasets=TASK_LIST_RETRIEVAL_ZH)
+            },
+            {
+                "language": "French",
+                "description": "**Retrieval French Leaderboard** 🔎🇫🇷",
+                "credits": french_credits,
+                "data": DATA_RETRIEVAL_FR,
+                "refresh": partial(get_mteb_data, tasks=["Retrieval"], datasets=TASK_LIST_RETRIEVAL_FR)
+            },
+            {
+                "language": "Polish",
+                "description": "**Retrieval Polish Leaderboard** 🔎🇵🇱",
+                "credits": polish_credits,
+                "data": DATA_RETRIEVAL_PL,
+                "refresh": partial(get_mteb_data, tasks=["Retrieval"], datasets=TASK_LIST_RETRIEVAL_PL)
+            }
+        ]
+    },
+    "STS": {
+        "metric": "Spearman correlation based on cosine similarity",
+        "data": [
+            {
+                "language": "English",
+                "description": "**STS English Leaderboard** 🤖",
+                "data": DATA_STS_EN,
+                "refresh": partial(get_mteb_data, tasks=["STS"], datasets=TASK_LIST_STS)
+            },
+            {
+                "language": "Chinese",
+                "description": "**STS Chinese Leaderboard** 🤖🇨🇳",
+                "credits": chinese_credits,
+                "data": DATA_STS_ZH,
+                "refresh": partial(get_mteb_data, tasks=["STS"], datasets=TASK_LIST_STS_ZH)
+            },
+            {
+                "language": "French",
+                "description": "**STS French Leaderboard** 🤖🇫🇷",
+                "credits": french_credits,
+                "data": DATA_STS_FR,
+                "refresh": partial(get_mteb_data, tasks=["STS"], datasets=TASK_LIST_STS_FR)
+            },
+            {
+                "language": "Polish",
+                "description": "**STS Polish Leaderboard** 🤖🇵🇱",
+                "credits": polish_credits,
+                "data": DATA_STS_PL,
+                "refresh": partial(get_mteb_data, tasks=["STS"], datasets=TASK_LIST_STS_PL)
+            },
+            {
+                "language": "Other",
+                "language_long": "Arabic, Chinese, Dutch, English, French, German, Italian, Korean, Polish, Russian, Spanish (Only language combos not included in the other tabs)",
+                "description": "**STS Other Leaderboard** 👽",
+                "data": DATA_STS_OTHER,
+                "refresh": partial(get_mteb_data, tasks=["STS"], datasets=TASK_LIST_STS_OTHER)
+            },
+        ]
+    },
+    "Summarization": {
+        "metric": "Spearman correlation	based on cosine similarity",
+        "data": [
+            {
+                "language": "English",
+                "description": "**Summarization Leaderboard** 📜",
+                "data": DATA_SUMMARIZATION,
+                "refresh": partial(get_mteb_data, tasks=TASK_LIST_SUMMARIZATION)
+            },
+            {
+                "language": "French",
+                "description": "**Summarization Leaderboard** 📜",
+                "credits": french_credits,
+                "data": DATA_SUMMARIZATION_FR,
+                "refresh": partial(get_mteb_data, tasks=TASK_LIST_SUMMARIZATION_FR)
+            }
+        ]
+    }
+}
+dataframes = []
+
+with gr.Blocks(css=css) as block:
     with gr.Tabs():
-        with gr.TabItem("Overall"):
-            with gr.TabItem("English"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Overall MTEB English leaderboard** 🔮
-                    
-                    - **Metric:** Various, refer to task tabs
-                    - **Languages:** English
-                    """)
-                with gr.Row():
-                    data_overall = gr.components.Dataframe(
-                        DATA_OVERALL,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_OVERALL.columns),
-                        type="pandas",
-                        height=600,
-                    )
-                with gr.Row():
-                    data_run_overall = gr.Button("Refresh")
-                    data_run_overall.click(get_mteb_average, inputs=None, outputs=data_overall)
-            with gr.TabItem("Chinese"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Overall MTEB Chinese leaderboard (C-MTEB)** 🔮🇨🇳
-                    
-                    - **Metric:** Various, refer to task tabs
-                    - **Languages:** Chinese
-                    - **Credits:** [FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding)
-                    """)
-                with gr.Row():
-                    data_overall_zh = gr.components.Dataframe(
-                        DATA_OVERALL_ZH,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_OVERALL_ZH.columns),
-                        type="pandas",
-                        height=600,
-                    )
-                with gr.Row():
-                    data_run_overall_zh = gr.Button("Refresh")
-                    data_run_overall_zh.click(get_mteb_average_zh, inputs=None, outputs=data_overall_zh)    
-            with gr.TabItem("French"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Overall MTEB French leaderboard (F-MTEB)** 🔮🇫🇷
-                    
-                    - **Metric:** Various, refer to task tabs
-                    - **Languages:** French
-                    - **Credits:** [Lyon-NLP](https://github.com/Lyon-NLP): [Gabriel Sequeira](https://github.com/GabrielSequeira), [Imene Kerboua](https://github.com/imenelydiaker), [Wissam Siblini](https://github.com/wissam-sib), [Mathieu Ciancone](https://github.com/MathieuCiancone), [Marion Schaeffer](https://github.com/schmarion)
-                    """)
-                with gr.Row():
-                    data_overall_fr = gr.components.Dataframe(
-                        DATA_OVERALL_FR,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_OVERALL_FR.columns),
-                        type="pandas",
-                        height=600,
-                    )
-                with gr.Row():
-                    data_overall_fr = gr.Button("Refresh")
-                    data_overall_fr.click(get_mteb_average_fr, inputs=None, outputs=data_overall_fr)                                  
-            with gr.TabItem("Polish"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Overall MTEB Polish leaderboard (PL-MTEB)** 🔮🇵🇱
-                    
-                    - **Metric:** Various, refer to task tabs
-                    - **Languages:** Polish
-                    - **Credits:** [Rafał Poświata](https://github.com/rafalposwiata), [Konrad Wojtasik](https://github.com/kwojtasi) & [BEIR-PL](https://arxiv.org/abs/2305.19840)
-                    """)
-                with gr.Row():
-                    data_overall_pl = gr.components.Dataframe(
-                        DATA_OVERALL_PL,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_OVERALL_PL.columns),
-                        type="pandas",
-                        height=600,                        
-                    )
-                with gr.Row():
-                    data_run_overall_pl = gr.Button("Refresh")
-                    data_run_overall_pl.click(get_mteb_average_pl, inputs=None, outputs=data_overall_pl)                    
-        with gr.TabItem("Bitext Mining"):
-            with gr.TabItem("English-X"):
-                with gr.Row():
-                        gr.Markdown("""
-                        **Bitext Mining English-X Leaderboard** 🎌
-                        
-                        - **Metric:** [F1](https://huggingface.co/spaces/evaluate-metric/f1)
-                        - **Languages:** 117 (Pairs of: English & other language)
-                        """)
-                with gr.Row():
-                    data_bitext_mining = gr.components.Dataframe(
-                        DATA_BITEXT_MINING,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_BITEXT_MINING.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_bitext_mining = gr.Button("Refresh")
-                    data_run_bitext_mining.click(
-                        partial(get_mteb_data, tasks=["BitextMining"], datasets=TASK_LIST_BITEXT_MINING),
-                        outputs=data_bitext_mining,
-                    )
-            with gr.TabItem("Danish"):
-                with gr.Row():
-                        gr.Markdown("""
-                        **Bitext Mining Danish Leaderboard** 🎌🇩🇰
-                        
-                        - **Metric:** [F1](https://huggingface.co/spaces/evaluate-metric/f1)
-                        - **Languages:** Danish & Bornholmsk (Danish Dialect)
-                        - **Credits:** [Kenneth Enevoldsen](https://github.com/KennethEnevoldsen), [scandinavian-embedding-benchmark](https://kennethenevoldsen.github.io/scandinavian-embedding-benchmark/)
-                        """)
-                with gr.Row():
-                    data_bitext_mining_da = gr.components.Dataframe(
-                        DATA_BITEXT_MINING_OTHER,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_BITEXT_MINING_OTHER.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_bitext_mining_da = gr.Button("Refresh")
-                    data_run_bitext_mining_da.click(
-                        partial(get_mteb_data, tasks=["BitextMining"], datasets=TASK_LIST_BITEXT_MINING_OTHER),
-                        outputs=data_bitext_mining_da,
-                    )
-        with gr.TabItem("Classification"):
-            with gr.TabItem("English"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Classification English Leaderboard** ❤️
-                    
-                    - **Metric:** [Accuracy](https://huggingface.co/spaces/evaluate-metric/accuracy)
-                    - **Languages:** English
-                    """)
-                with gr.Row():
-                    data_classification_en = gr.components.Dataframe(
-                        DATA_CLASSIFICATION_EN,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_CLASSIFICATION_EN.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_classification_en = gr.Button("Refresh")
-                    data_run_classification_en.click(
-                        partial(get_mteb_data, tasks=["Classification"], langs=["en"]),
-                        outputs=data_classification_en,
-                    )
-            with gr.TabItem("Chinese"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Classification Chinese Leaderboard** 🧡🇨🇳
-                    
-                    - **Metric:** [Accuracy](https://huggingface.co/spaces/evaluate-metric/accuracy)
-                    - **Languages:** Chinese
-                    - **Credits:** [FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding)
-                    """)
-                with gr.Row():
-                    data_classification_zh = gr.components.Dataframe(
-                        DATA_CLASSIFICATION_ZH,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_CLASSIFICATION_ZH.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_classification_zh = gr.Button("Refresh")
-                    data_run_classification_zh.click(
-                        partial(get_mteb_data, tasks=["Classification"], datasets=TASK_LIST_CLASSIFICATION_ZH),
-                        outputs=data_classification_zh,
-                    )
-            with gr.TabItem("Danish"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Classification Danish Leaderboard** 🤍🇩🇰
-                    
-                    - **Metric:** [Accuracy](https://huggingface.co/spaces/evaluate-metric/accuracy)
-                    - **Languages:** Danish
-                    - **Credits:** [Kenneth Enevoldsen](https://github.com/KennethEnevoldsen), [scandinavian-embedding-benchmark](https://kennethenevoldsen.github.io/scandinavian-embedding-benchmark/)
-                    """)
-                with gr.Row():
-                    data_classification_da = gr.components.Dataframe(
-                        DATA_CLASSIFICATION_DA,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_CLASSIFICATION_DA.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_classification_da = gr.Button("Refresh")
-                    data_run_classification_da.click(
-                        partial(get_mteb_data, tasks=["Classification"], datasets=TASK_LIST_CLASSIFICATION_DA),
-                        outputs=data_run_classification_da,
-                    )
-            with gr.TabItem("French"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Classification French Leaderboard** 💙🇫🇷
-                    
-                    - **Metric:** [Accuracy](https://huggingface.co/spaces/evaluate-metric/accuracy)
-                    - **Languages:** French
-                    - **Credits:** [Lyon-NLP](https://github.com/Lyon-NLP): [Gabriel Sequeira](https://github.com/GabrielSequeira), [Imene Kerboua](https://github.com/imenelydiaker), [wissam-sib](https://github.com/wissam-sib), [Mathieu Ciancone](https://github.com/MathieuCiancone), [Marion Schaeffer](https://github.com/schmarion)
-                    """)
-                with gr.Row():
-                    data_classification_fr = gr.components.Dataframe(
-                        DATA_CLASSIFICATION_FR,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_CLASSIFICATION_FR.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_classification_fr = gr.Button("Refresh")
-                    data_run_classification_fr.click(
-                        partial(get_mteb_data, tasks=["Classification"], datasets=TASK_LIST_CLASSIFICATION_FR),
-                        outputs=data_run_classification_fr,
-                    )                    
-            with gr.TabItem("Norwegian"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Classification Norwegian Leaderboard** 💙🇳🇴
-                    
-                    - **Metric:** [Accuracy](https://huggingface.co/spaces/evaluate-metric/accuracy)
-                    - **Languages:** Norwegian Bokmål
-                    - **Credits:** [Kenneth Enevoldsen](https://github.com/KennethEnevoldsen), [scandinavian-embedding-benchmark](https://kennethenevoldsen.github.io/scandinavian-embedding-benchmark/)                                
-                    """)
-                with gr.Row():
-                    data_classification_nb = gr.components.Dataframe(
-                        DATA_CLASSIFICATION_NB,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_CLASSIFICATION_NB.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_classification_nb = gr.Button("Refresh")
-                    data_run_classification_nb.click(
-                        partial(get_mteb_data, tasks=["Classification"], datasets=TASK_LIST_CLASSIFICATION_NB),
-                        outputs=data_classification_nb,
-                    )
-            with gr.TabItem("Polish"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Classification Polish Leaderboard** 🤍🇵🇱
-                    
-                    - **Metric:** [Accuracy](https://huggingface.co/spaces/evaluate-metric/accuracy)
-                    - **Languages:** Polish
-                    - **Credits:** [Rafał Poświata](https://github.com/rafalposwiata)
-                    """)
-                with gr.Row():
-                    data_classification_pl = gr.components.Dataframe(
-                        DATA_CLASSIFICATION_PL,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_CLASSIFICATION_PL.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_classification_pl = gr.Button("Refresh")
-                    data_run_classification_pl.click(
-                        partial(get_mteb_data, tasks=["Classification"], datasets=TASK_LIST_CLASSIFICATION_PL),
-                        outputs=data_classification_pl,
-                    )
-            with gr.TabItem("Swedish"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Classification Swedish Leaderboard** 💛🇸🇪
-                    
-                    - **Metric:** [Accuracy](https://huggingface.co/spaces/evaluate-metric/accuracy)
-                    - **Languages:** Swedish
-                    - **Credits:** [Kenneth Enevoldsen](https://github.com/KennethEnevoldsen), [scandinavian-embedding-benchmark](https://kennethenevoldsen.github.io/scandinavian-embedding-benchmark/)
-                    """)
-                with gr.Row():
-                    data_classification_sv = gr.components.Dataframe(
-                        DATA_CLASSIFICATION_SV,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_CLASSIFICATION_SV.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_classification_sv = gr.Button("Refresh")
-                    data_run_classification_sv.click(
-                        partial(get_mteb_data, tasks=["Classification"], datasets=TASK_LIST_CLASSIFICATION_SV),
-                        outputs=data_classification_sv,
-                    )
-            with gr.TabItem("Other"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Classification Other Languages Leaderboard** 💜💚💙
-                    
-                    - **Metric:** [Accuracy](https://huggingface.co/spaces/evaluate-metric/accuracy)
-                    - **Languages:** 47 (Only languages not included in the other tabs)
-                    """)
-                with gr.Row():
-                    data_classification = gr.components.Dataframe(
-                        DATA_CLASSIFICATION_OTHER,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_CLASSIFICATION_OTHER) * 10,
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_classification = gr.Button("Refresh")
-                    data_run_classification.click(
-                        partial(get_mteb_data, tasks=["Classification"], datasets=TASK_LIST_CLASSIFICATION_OTHER),
-                        outputs=data_classification,
-                    )                                    
-        with gr.TabItem("Clustering"):
-            with gr.TabItem("English"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Clustering Leaderboard** ✨
-                    
-                    - **Metric:** Validity Measure (v_measure)
-                    - **Languages:** English
-                    """)
-                with gr.Row():
-                    data_clustering = gr.components.Dataframe(
-                        DATA_CLUSTERING,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_CLUSTERING.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_clustering_en = gr.Button("Refresh")
-                    data_run_clustering_en.click(
-                        partial(get_mteb_data, tasks=["Clustering"], datasets=TASK_LIST_CLUSTERING),
-                        outputs=data_clustering,
-                    )
-            with gr.TabItem("Chinese"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Clustering Chinese Leaderboard** ✨🇨🇳
-                    
-                    - **Metric:** Validity Measure (v_measure)
-                    - **Languages:** Chinese
-                    - **Credits:** [FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding)
-                    """)
-                with gr.Row():
-                    data_clustering_zh = gr.components.Dataframe(
-                        DATA_CLUSTERING_ZH,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_CLUSTERING_ZH.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_clustering_zh = gr.Button("Refresh")
-                    data_run_clustering_zh.click(
-                        partial(get_mteb_data, tasks=["Clustering"], datasets=TASK_LIST_CLUSTERING_ZH),
-                        outputs=data_clustering_zh,
-                    )
-            with gr.TabItem("French"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Clustering French Leaderboard** ✨🇫🇷
-                    
-                    - **Metric:** Validity Measure (v_measure)
-                    - **Languages:** French
-                    - **Credits:** [Lyon-NLP](https://github.com/Lyon-NLP): [Gabriel Sequeira](https://github.com/GabrielSequeira), [Imene Kerboua](https://github.com/imenelydiaker), [wissam-sib](https://github.com/wissam-sib), [Mathieu Ciancone](https://github.com/MathieuCiancone), [Marion Schaeffer](https://github.com/schmarion)
-                    """)
-                with gr.Row():
-                    data_clustering_fr = gr.components.Dataframe(
-                        DATA_CLUSTERING_FR,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_CLUSTERING_FR.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_clustering_fr = gr.Button("Refresh")
-                    data_run_clustering_fr.click(
-                        partial(get_mteb_data, tasks=["Clustering"], datasets=TASK_LIST_CLUSTERING_FR),
-                        outputs=data_clustering_fr,
-                    )
-            with gr.TabItem("German"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Clustering German Leaderboard** ✨🇩🇪
-                    
-                    - **Metric:** Validity Measure (v_measure)
-                    - **Languages:** German
-                    - **Credits:** [Silvan](https://github.com/slvnwhrl)
-                    """)
-                with gr.Row():
-                    data_clustering_de = gr.components.Dataframe(
-                        DATA_CLUSTERING_DE,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_CLUSTERING_DE.columns) * 2,
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_clustering_de = gr.Button("Refresh")
-                    data_run_clustering_de.click(
-                        partial(get_mteb_data, tasks=["Clustering"], datasets=TASK_LIST_CLUSTERING_DE),
-                        outputs=data_clustering_de,
-                    )
-            with gr.TabItem("Polish"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Clustering Polish Leaderboard** ✨🇵🇱
-                    
-                    - **Metric:** Validity Measure (v_measure)
-                    - **Languages:** Polish
-                    - **Credits:** [Rafał Poświata](https://github.com/rafalposwiata)
-                    """)
-                with gr.Row():
-                    data_clustering_pl = gr.components.Dataframe(
-                        DATA_CLUSTERING_PL,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_CLUSTERING_PL.columns) * 2,
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_clustering_pl = gr.Button("Refresh")
-                    data_run_clustering_pl.click(
-                        partial(get_mteb_data, tasks=["Clustering"], datasets=TASK_LIST_CLUSTERING_PL),
-                        outputs=data_clustering_pl,
-                    )
-        with gr.TabItem("Pair Classification"):
-            with gr.TabItem("English"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Pair Classification English Leaderboard** 🎭
-                    
-                    - **Metric:** Average Precision based on Cosine Similarities (cos_sim_ap)
-                    - **Languages:** English
-                    """)
-                with gr.Row():
-                    data_pair_classification = gr.components.Dataframe(
-                        DATA_PAIR_CLASSIFICATION,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_PAIR_CLASSIFICATION.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_pair_classification = gr.Button("Refresh")
-                    data_run_pair_classification.click(
-                        partial(get_mteb_data, tasks=["PairClassification"], datasets=TASK_LIST_PAIR_CLASSIFICATION),
-                        outputs=data_pair_classification,
-                    )
-            with gr.TabItem("Chinese"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Pair Classification Chinese Leaderboard** 🎭🇨🇳
-                    
-                    - **Metric:** Average Precision based on Cosine Similarities (cos_sim_ap)
-                    - **Languages:** Chinese
-                    - **Credits:** [FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding)
-                    """)
-                with gr.Row():
-                    data_pair_classification_zh = gr.components.Dataframe(
-                        DATA_PAIR_CLASSIFICATION_ZH,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_PAIR_CLASSIFICATION_ZH.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_pair_classification_zh = gr.Button("Refresh")
-                    data_run_pair_classification_zh.click(
-                        partial(get_mteb_data, tasks=["PairClassification"], datasets=TASK_LIST_PAIR_CLASSIFICATION_ZH),
-                        outputs=data_pair_classification_zh,
-                    )
-            with gr.TabItem("French"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Pair Classification French Leaderboard** 🎭🇫🇷
-                    
-                    - **Metric:** Average Precision based on Cosine Similarities (cos_sim_ap)
-                    - **Languages:** French
-                    - **Credits:** [Lyon-NLP](https://github.com/Lyon-NLP): [Gabriel Sequeira](https://github.com/GabrielSequeira), [Imene Kerboua](https://github.com/imenelydiaker), [wissam-sib](https://github.com/wissam-sib), [Mathieu Ciancone](https://github.com/MathieuCiancone), [Marion Schaeffer](https://github.com/schmarion)
-                    """)
-                with gr.Row():
-                    data_pair_classification_fr = gr.components.Dataframe(
-                        DATA_PAIR_CLASSIFICATION_FR,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_PAIR_CLASSIFICATION_FR.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_pair_classification_fr = gr.Button("Refresh")
-                    data_run_pair_classification_fr.click(
-                        partial(get_mteb_data, tasks=["PairClassification"], datasets=TASK_LIST_PAIR_CLASSIFICATION_FR),
-                        outputs=data_pair_classification_fr,
-                    )
-            with gr.TabItem("Polish"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Pair Classification Polish Leaderboard** 🎭🇵🇱
-                    
-                    - **Metric:** Average Precision based on Cosine Similarities (cos_sim_ap)
-                    - **Languages:** Polish
-                    - **Credits:** [Rafał Poświata](https://github.com/rafalposwiata)
-                    """)
-                with gr.Row():
-                    data_pair_classification_pl = gr.components.Dataframe(
-                        DATA_PAIR_CLASSIFICATION_PL,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_PAIR_CLASSIFICATION_PL.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_pair_classification_pl = gr.Button("Refresh")
-                    data_run_pair_classification_pl.click(
-                        partial(get_mteb_data, tasks=["PairClassification"], datasets=TASK_LIST_PAIR_CLASSIFICATION_PL),
-                        outputs=data_pair_classification_pl,
-                    )               
-        with gr.TabItem("Reranking"):
-            with gr.TabItem("English"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Reranking English Leaderboard** 🥈
-                    
-                    - **Metric:** Mean Average Precision (MAP)
-                    - **Languages:** English
-                    """)
-                with gr.Row():
-                    data_reranking = gr.components.Dataframe(
-                        DATA_RERANKING,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_RERANKING.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_reranking = gr.Button("Refresh")
-                    data_run_reranking.click(
-                        partial(get_mteb_data, tasks=["Reranking"], datasets=TASK_LIST_RERANKING),
-                        outputs=data_reranking,
-                    )
-            with gr.TabItem("Chinese"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Reranking Chinese Leaderboard** 🥈🇨🇳
-                    
-                    - **Metric:** Mean Average Precision (MAP)
-                    - **Languages:** Chinese
-                    - **Credits:** [FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding)
-                    """)
-                with gr.Row():
-                    data_reranking_zh = gr.components.Dataframe(
-                        DATA_RERANKING_ZH,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_RERANKING_ZH.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_reranking_zh = gr.Button("Refresh")
-                    data_run_reranking_zh.click(
-                        partial(get_mteb_data, tasks=["Reranking"], datasets=TASK_LIST_RERANKING_ZH),
-                        outputs=data_reranking_zh,
-                    )
-            with gr.TabItem("French"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Reranking French Leaderboard** 🥈🇫🇷
-                    
-                    - **Metric:** Mean Average Precision (MAP)
-                    - **Languages:** French
-                    - **Credits:** [Lyon-NLP](https://github.com/Lyon-NLP): [Gabriel Sequeira](https://github.com/GabrielSequeira), [Imene Kerboua](https://github.com/imenelydiaker), [wissam-sib](https://github.com/wissam-sib), [Mathieu Ciancone](https://github.com/MathieuCiancone), [Marion Schaeffer](https://github.com/schmarion)
-                    """)
-                with gr.Row():
-                    data_reranking_fr = gr.components.Dataframe(
-                        DATA_RERANKING_FR,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_RERANKING_FR.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_reranking_fr = gr.Button("Refresh")
-                    data_run_reranking_fr.click(
-                        partial(get_mteb_data, tasks=["Reranking"], datasets=TASK_LIST_RERANKING_FR),
-                        outputs=data_reranking_fr,
-                    )
-        with gr.TabItem("Retrieval"):
-            with gr.TabItem("English"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Retrieval English Leaderboard** 🔎
-                    
-                    - **Metric:** Normalized Discounted Cumulative Gain @ k (ndcg_at_10)
-                    - **Languages:** English
-                    """)
-                with gr.Row():
-                    data_retrieval = gr.components.Dataframe(
-                        DATA_RETRIEVAL,
-                        # Add support for more columns than existing as a buffer for CQADupstack & other Retrieval tasks (e.g. MSMARCOv2)
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_RETRIEVAL.columns) * 2,
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_retrieval = gr.Button("Refresh")
-                    data_run_retrieval.click(
-                        partial(get_mteb_data, tasks=["Retrieval"], datasets=TASK_LIST_RETRIEVAL),
-                        outputs=data_retrieval,
-                    )
-            with gr.TabItem("Chinese"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Retrieval Chinese Leaderboard** 🔎🇨🇳
-                    
-                    - **Metric:** Normalized Discounted Cumulative Gain @ k (ndcg_at_10)
-                    - **Languages:** Chinese
-                    - **Credits:** [FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding)
-                    """)
-                with gr.Row():
-                    data_retrieval_zh = gr.components.Dataframe(
-                        DATA_RETRIEVAL_ZH,
-                        # Add support for more columns than existing as a buffer for CQADupstack & other Retrieval tasks (e.g. MSMARCOv2)
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_RETRIEVAL_ZH.columns) * 2,
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_retrieval_zh = gr.Button("Refresh")
-                    data_run_retrieval_zh.click(
-                        partial(get_mteb_data, tasks=["Retrieval"], datasets=TASK_LIST_RETRIEVAL_ZH),
-                        outputs=data_retrieval_zh,
-                    )
-            with gr.TabItem("French"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Retrieval French Leaderboard** 🔎🇫🇷
-                    
-                    - **Metric:** Normalized Discounted Cumulative Gain @ k (ndcg_at_10)
-                    - **Languages:** French
-                    - **Credits:** [Lyon-NLP](https://github.com/Lyon-NLP): [Gabriel Sequeira](https://github.com/GabrielSequeira), [Imene Kerboua](https://github.com/imenelydiaker), [wissam-sib](https://github.com/wissam-sib), [Mathieu Ciancone](https://github.com/MathieuCiancone), [Marion Schaeffer](https://github.com/schmarion)
-                    """)
-                with gr.Row():
-                    data_retrieval_fr = gr.components.Dataframe(
-                        DATA_RETRIEVAL_FR,
-                        # Add support for more columns than existing as a buffer for CQADupstack & other Retrieval tasks (e.g. MSMARCOv2)
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_RETRIEVAL_FR.columns) * 2,
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_retrieval_fr = gr.Button("Refresh")
-                    data_run_retrieval_fr.click(
-                        partial(get_mteb_data, tasks=["Retrieval"], datasets=TASK_LIST_RETRIEVAL_FR),
-                        outputs=data_retrieval_fr,
-                    )                    
-            with gr.TabItem("Polish"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Retrieval Polish Leaderboard** 🔎🇵🇱
-                    
-                    - **Metric:** Normalized Discounted Cumulative Gain @ k (ndcg_at_10)
-                    - **Languages:** Polish
-                    - **Credits:** [Konrad Wojtasik](https://github.com/kwojtasi) & [BEIR-PL](https://arxiv.org/abs/2305.19840)
-                    """)
-                with gr.Row():
-                    data_retrieval_pl = gr.components.Dataframe(
-                        DATA_RETRIEVAL_PL,
-                        # Add support for more columns than existing as a buffer for CQADupstack & other Retrieval tasks (e.g. MSMARCOv2)
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_RETRIEVAL_PL.columns) * 2,
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_retrieval_pl = gr.Button("Refresh")
-                    data_run_retrieval_pl.click(
-                        partial(get_mteb_data, tasks=["Retrieval"], datasets=TASK_LIST_RETRIEVAL_PL),
-                        outputs=data_retrieval_pl,
-                    )
-        with gr.TabItem("STS"):
-            with gr.TabItem("English"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **STS English Leaderboard** 🤖
-                    
-                    - **Metric:** Spearman correlation based on cosine similarity
-                    - **Languages:** English
-                    """)
-                with gr.Row():
-                    data_sts_en = gr.components.Dataframe(
-                        DATA_STS_EN,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_STS_EN.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_sts_en = gr.Button("Refresh")
-                    data_run_sts_en.click(
-                        partial(get_mteb_data, tasks=["STS"], datasets=TASK_LIST_STS),
-                        outputs=data_sts_en,
-                    )
-            with gr.TabItem("Chinese"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **STS Chinese Leaderboard** 🤖🇨🇳
-                    
-                    - **Metric:** Spearman correlation based on cosine similarity
-                    - **Languages:** Chinese
-                    - **Credits:** [FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding)
-                    """)
-                with gr.Row():
-                    data_sts_zh = gr.components.Dataframe(
-                        DATA_STS_ZH,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_STS_ZH.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_sts_zh = gr.Button("Refresh")
-                    data_run_sts_zh.click(
-                        partial(get_mteb_data, tasks=["STS"], datasets=TASK_LIST_STS_ZH),
-                        outputs=data_sts_zh,
-                    )
-            with gr.TabItem("French"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **STS French Leaderboard** 🤖🇫🇷
-                    
-                    - **Metric:** Spearman correlation based on cosine similarity
-                    - **Languages:** French
-                    - **Credits:** [Lyon-NLP](https://github.com/Lyon-NLP): [Gabriel Sequeira](https://github.com/GabrielSequeira), [Imene Kerboua](https://github.com/imenelydiaker), [wissam-sib](https://github.com/wissam-sib), [Mathieu Ciancone](https://github.com/MathieuCiancone), [Marion Schaeffer](https://github.com/schmarion)
-                    """)
-                with gr.Row():
-                    data_sts_fr = gr.components.Dataframe(
-                        DATA_STS_FR,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_STS_FR.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_sts_fr = gr.Button("Refresh")
-                    data_run_sts_fr.click(
-                        partial(get_mteb_data, tasks=["STS"], datasets=TASK_LIST_STS_FR),
-                        outputs=data_sts_fr,
-                    )                    
-            with gr.TabItem("Polish"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **STS Polish Leaderboard** 🤖🇵🇱
-                    
-                    - **Metric:** Spearman correlation based on cosine similarity
-                    - **Languages:** Polish
-                    - **Credits:** [Rafał Poświata](https://github.com/rafalposwiata)
-                    """)
-                with gr.Row():
-                    data_sts_pl = gr.components.Dataframe(
-                        DATA_STS_PL,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_STS_PL.columns),
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_sts_pl = gr.Button("Refresh")
-                    data_run_sts_pl.click(
-                        partial(get_mteb_data, tasks=["STS"], datasets=TASK_LIST_STS_PL),
-                        outputs=data_sts_pl,
-                    )
-            with gr.TabItem("Other"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **STS Other Leaderboard** 👽
-                    
-                    - **Metric:** Spearman correlation based on cosine similarity
-                    - **Languages:** Arabic, Chinese, Dutch, English, French, German, Italian, Korean, Polish, Russian, Spanish (Only language combos not included in the other tabs)
-                    """)
-                with gr.Row():
-                    data_sts_other = gr.components.Dataframe(
-                        DATA_STS_OTHER,
-                        datatype=["number", "markdown"] + ["number"] * len(DATA_STS_OTHER.columns) * 2,
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_sts_other = gr.Button("Refresh")
-                    data_run_sts_other.click(
-                        partial(get_mteb_data, tasks=["STS"], datasets=TASK_LIST_STS_OTHER),
-                        outputs=data_sts_other,
-                    )
-        with gr.TabItem("Summarization"):
-            with gr.TabItem("English"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Summarization Leaderboard** 📜
-                    
-                    - **Metric:** Spearman correlation based on cosine similarity
-                    - **Languages:** English
-                    """)
-                with gr.Row():
-                    data_summarization = gr.components.Dataframe(
-                        DATA_SUMMARIZATION,
-                        datatype=["number", "markdown"] + ["number"] * 2,
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run = gr.Button("Refresh")
-                    data_run.click(
-                        partial(get_mteb_data, tasks=TASK_LIST_SUMMARIZATION),
-                        outputs=data_summarization,
-                    )
-            with gr.TabItem("French"):
-                with gr.Row():
-                    gr.Markdown("""
-                    **Summarization Leaderboard** 📜
-                    
-                    - **Metric:** Spearman correlation based on cosine similarity
-                    - **Languages:** French
-                    - **Credits:** [Lyon-NLP](https://github.com/Lyon-NLP): [Gabriel Sequeira](https://github.com/GabrielSequeira), [Imene Kerboua](https://github.com/imenelydiaker), [wissam-sib](https://github.com/wissam-sib), [Mathieu Ciancone](https://github.com/MathieuCiancone), [Marion Schaeffer](https://github.com/schmarion)                            
-                    """)
-                with gr.Row():
-                    data_summarization_fr = gr.components.Dataframe(
-                        DATA_SUMMARIZATION_FR,
-                        datatype=["number", "markdown"] + ["number"] * 2,
-                        type="pandas",
-                    )
-                with gr.Row():
-                    data_run_summarization_fr = gr.Button("Refresh")
-                    data_run_summarization_fr.click(
-                        partial(get_mteb_data, tasks=TASK_LIST_SUMMARIZATION_FR),
-                        outputs=data_run_summarization_fr,
-                    )
+        for task, task_values in data.items():
+            metric = task_values["metric"]
+            with gr.Tab(task):
+                for item in task_values["data"]:
+                    with gr.Tab(item["language"]):
+                        with gr.Row():
+                            gr.Markdown(f"""
+                            {item['description']}
+
+                            - **Metric:** {metric}
+                            - **Languages:** {item['language_long'] if 'language_long' in item else item['language']}
+                            {"- **Credits:** " + item['credits'] if "credits" in item else ''}
+                            """)
+                        with gr.Row():
+                            datatype = ["number", "markdown"] + ["number"] * len(item["data"])
+                            dataframe = gr.Dataframe(item["data"], datatype=datatype, type="pandas", height=600)
+                            dataframes.append(dataframe)
+                        with gr.Row():
+                            refresh_button = gr.Button("Refresh")
+                            refresh_button.click(item["refresh"], inputs=None, outputs=dataframe)
+
     gr.Markdown(f"""
     - **Total Datasets**: {NUM_DATASETS}
     - **Total Languages**: 113
@@ -2302,16 +1865,10 @@ with block:
     }
     ```
     """)
-    # Running the functions on page load in addition to when the button is clicked
-    # This is optional - If deactivated the data loaded at "Build time" is shown like for Overall tab
-    """
-    block.load(get_mteb_data, inputs=[task_bitext_mining], outputs=data_bitext_mining)
-    """
 
 block.queue(max_size=10)
 block.launch()
 
-
 # Possible changes:
 # Could add graphs / other visual content
 # Could add verification marks