fix: set use_instructions to True in models using prompts (#1616)

feat: set `use_instructions` to True in models using prompts

mteb/models/arctic_models.py

@@ -100,7 +100,7 @@
     license="apache-2.0",
     reference="https://huggingface.co/Snowflake/snowflake-arctic-embed-xs",
     similarity_fn_name="cosine",
-    use_instructions=False,
+    use_instructions=True,
     adapted_from="sentence-transformers/all-MiniLM-L6-v2",
     superseded_by=None,
 )
@@ -125,7 +125,7 @@
     license="apache-2.0",
     reference="https://huggingface.co/Snowflake/snowflake-arctic-embed-s",
     similarity_fn_name="cosine",
-    use_instructions=False,
+    use_instructions=True,
     adapted_from="intfloat/e5-small-unsupervised",
     superseded_by=None,
 )
@@ -150,7 +150,7 @@
     license="apache-2.0",
     reference="https://huggingface.co/Snowflake/snowflake-arctic-embed-m",
     similarity_fn_name="cosine",
-    use_instructions=False,
+    use_instructions=True,
     adapted_from="intfloat/e5-base-unsupervised",
     superseded_by="Snowflake/snowflake-arctic-embed-m-v1.5",
 )
@@ -174,7 +174,7 @@
     license="apache-2.0",
     reference="https://huggingface.co/Snowflake/snowflake-arctic-embed-m-long",
     similarity_fn_name="cosine",
-    use_instructions=False,
+    use_instructions=True,
     adapted_from="nomic-ai/nomic-embed-text-v1-unsupervised",
     superseded_by="Snowflake/snowflake-arctic-embed-m-v2.0",
 )
@@ -198,7 +198,7 @@
     license="apache-2.0",
     reference="https://huggingface.co/Snowflake/snowflake-arctic-embed-l",
     similarity_fn_name="cosine",
-    use_instructions=False,
+    use_instructions=True,
     adapted_from="intfloat/e5-base-unsupervised",
     superseded_by="Snowflake/snowflake-arctic-embed-l-v2.0",
 )
@@ -225,7 +225,7 @@
     license="apache-2.0",
     reference="https://huggingface.co/Snowflake/snowflake-arctic-embed-m-v1.5",
     similarity_fn_name="cosine",
-    use_instructions=False,
+    use_instructions=True,
     adapted_from=None,
     superseded_by="Snowflake/snowflake-arctic-embed-m-v2.0",
 )
@@ -249,7 +249,7 @@
     license="apache-2.0",
     reference="https://huggingface.co/Snowflake/snowflake-arctic-embed-m-v2.0",
     similarity_fn_name="cosine",
-    use_instructions=False,
+    use_instructions=True,
     adapted_from="Alibaba-NLP/gte-multilingual-base",
     superseded_by=None,
 )
@@ -273,7 +273,7 @@
     license="apache-2.0",
     reference="https://huggingface.co/Snowflake/snowflake-arctic-embed-l-v2.0",
     similarity_fn_name="cosine",
-    use_instructions=False,
+    use_instructions=True,
     adapted_from="BAAI/bge-m3-retromae",
     superseded_by=None,
 )

mteb/models/bge_models.py

@@ -26,7 +26,7 @@
     reference="https://huggingface.co/BAAI/bge-small-en-v1.5",
     similarity_fn_name="cosine",
     framework=["Sentence Transformers", "PyTorch"],
-    use_instructions=False,
+    use_instructions=True,
 )
 
 bge_base_en_v1_5 = ModelMeta(
@@ -49,7 +49,7 @@
     reference="https://huggingface.co/BAAI/bge-base-en-v1.5",
     similarity_fn_name="cosine",
     framework=["Sentence Transformers", "PyTorch"],
-    use_instructions=False,
+    use_instructions=True,
 )
 
 bge_large_en_v1_5 = ModelMeta(
@@ -72,5 +72,5 @@
     reference="https://huggingface.co/BAAI/bge-large-en-v1.5",
     similarity_fn_name="cosine",
     framework=["Sentence Transformers", "PyTorch"],
-    use_instructions=False,
+    use_instructions=True,
 )

mteb/models/cohere_models.py

@@ -234,7 +234,7 @@ def encode(
     license=None,
     similarity_fn_name="cosine",
     framework=["API"],
-    use_instructions=False,
+    use_instructions=True,
 )
 
 cohere_eng_3 = ModelMeta(
@@ -256,7 +256,7 @@ def encode(
     license=None,
     similarity_fn_name="cosine",
     framework=["API"],
-    use_instructions=False,
+    use_instructions=True,
 )
 
 cohere_mult_light_3 = ModelMeta(
@@ -278,7 +278,7 @@ def encode(
     license=None,
     similarity_fn_name="cosine",
     framework=["API"],
-    use_instructions=False,
+    use_instructions=True,
 )
 
 cohere_eng_light_3 = ModelMeta(
@@ -300,5 +300,5 @@ def encode(
     license=None,
     similarity_fn_name="cosine",
     framework=["API"],
-    use_instructions=False,
+    use_instructions=True,
 )

mteb/models/jina_models.py

@@ -220,6 +220,6 @@ def encode(
     license="cc-by-nc-4.0",
     similarity_fn_name="cosine",
     framework=["Sentence Transformers", "PyTorch"],
-    use_instructions=False,
+    use_instructions=True,
     reference="https://huggingface.co/jinaai/jina-embeddings-v3",
 )

mteb/models/ru_sentence_models.py

@@ -79,7 +79,7 @@
         sentence_transformers_loader,
         model_name="deepvk/USER-base",
         revision="436a489a2087d61aa670b3496a9915f84e46c861",
-        prompts={"query": "query: ", "passage": "passage: "},
+        model_prompts={"query": "query: ", "passage": "passage: "},
     ),
     name="deepvk/USER-base",
     languages=["rus_Cyrl"],
@@ -94,7 +94,7 @@
     reference="https://huggingface.co/ai-forever/sbert_large_mt_nlu_ru",
     similarity_fn_name="cosine",
     framework=["Sentence Transformers", "PyTorch"],
-    use_instructions=False,
+    use_instructions=True,
 )
 
 deberta_v1_ru = ModelMeta(
@@ -234,4 +234,5 @@
     open_weights=True,
     revision="89fb1651989adbb1cfcfdedafd7d102951ad0555",
     release_date="2024-07-29",
+    use_instructions=True,
 )

mteb/models/uae_models.py

@@ -74,5 +74,5 @@ def encode(
     similarity_fn_name="cosine",
     framework=["Sentence Transformers", "PyTorch"],
     reference="https://huggingface.co/WhereIsAI/UAE-Large-V1",
-    use_instructions=False,
+    use_instructions=True,
 )

mteb/models/voyage_models.py

@@ -181,7 +181,7 @@ def _batched_encode(
     reference="https://blog.voyageai.com/2024/06/03/domain-specific-embeddings-finance-edition-voyage-finance-2/",
     similarity_fn_name="cosine",
     framework=["API"],
-    use_instructions=False,
+    use_instructions=True,
 )
 
 voyage_law_2 = ModelMeta(
@@ -203,7 +203,7 @@ def _batched_encode(
     reference="https://blog.voyageai.com/2024/04/15/domain-specific-embeddings-and-retrieval-legal-edition-voyage-law-2/",
     similarity_fn_name="cosine",
     framework=["API"],
-    use_instructions=False,
+    use_instructions=True,
 )
 
 voyage_code_2 = ModelMeta(
@@ -225,7 +225,7 @@ def _batched_encode(
     reference="https://blog.voyageai.com/2024/01/23/voyage-code-2-elevate-your-code-retrieval/",
     similarity_fn_name="cosine",
     framework=["API"],
-    use_instructions=False,
+    use_instructions=True,
 )
 
 voyage_large_2 = ModelMeta(
@@ -247,7 +247,7 @@ def _batched_encode(
     reference="https://blog.voyageai.com/2023/10/29/voyage-embeddings/",
     similarity_fn_name="cosine",
     framework=["API"],
-    use_instructions=False,
+    use_instructions=True,
 )
 
 voyage_2 = ModelMeta(
@@ -269,7 +269,7 @@ def _batched_encode(
     reference="https://blog.voyageai.com/2023/10/29/voyage-embeddings/",
     similarity_fn_name="cosine",
     framework=["API"],
-    use_instructions=False,
+    use_instructions=True,
 )
 voyage_multilingual_2 = ModelMeta(
     name="voyageai/voyage-multilingual-2",
@@ -290,7 +290,7 @@ def _batched_encode(
     reference="https://blog.voyageai.com/2024/06/10/voyage-multilingual-2-multilingual-embedding-model/",
     similarity_fn_name="cosine",
     framework=["API"],
-    use_instructions=False,
+    use_instructions=True,
 )
 
 voyage_3 = ModelMeta(
@@ -312,7 +312,7 @@ def _batched_encode(
     reference="https://blog.voyageai.com/2024/09/18/voyage-3/",
     similarity_fn_name="cosine",
     framework=["API"],
-    use_instructions=False,
+    use_instructions=True,
 )
 
 voyage_3_lite = ModelMeta(
@@ -334,5 +334,5 @@ def _batched_encode(
     reference="https://blog.voyageai.com/2024/09/18/voyage-3/",
     similarity_fn_name="cosine",
     framework=["API"],
-    use_instructions=False,
+    use_instructions=True,
 )