February freshness, part 1

Author: HeidiSteen

articles/search/cognitive-search-concept-intro.md

@@ -10,14 +10,14 @@ ms.service: azure-ai-search
 ms.custom:
   - ignite-2023
 ms.topic: conceptual
-ms.date: 09/04/2024
+ms.date: 02/24/2025
 ---
 
 # AI enrichment in Azure AI Search
 
 In Azure AI Search, *AI enrichment* refers to integration with [Azure AI services](/azure/ai-services/what-are-ai-services) to process content that isn't searchable in its raw form. Through enrichment, analysis and inference are used to create searchable content and structure where none previously existed. 
 
-Because Azure AI Search is used for text and vector queries, the purpose of AI enrichment is to improve the utility of your content in search-related scenarios. Raw content must be text or images (you can't enrich vectors), but the output of an enrichment pipeline can be vectorized and indexed in a vector index using skills like [Text Split skill](cognitive-search-skill-textsplit.md) for chunking and [AzureOpenAIEmbedding skill](cognitive-search-skill-azure-openai-embedding.md) for encoding. For more information about using skills in vector scenarios, see [Integrated data chunking and embedding](vector-search-integrated-vectorization.md).
+Because Azure AI Search is used for text and vector queries, the purpose of AI enrichment is to improve the utility of your content in search-related scenarios. Raw content must be text or images (you can't enrich vectors), but the output of an enrichment pipeline can be vectorized and indexed in a search index using skills like [Text Split skill](cognitive-search-skill-textsplit.md) for chunking and [AzureOpenAIEmbedding skill](cognitive-search-skill-azure-openai-embedding.md) for vector encoding. For more information about using skills in vector scenarios, see [Integrated data chunking and embedding](vector-search-integrated-vectorization.md).
 
 AI enrichment is based on [*skills*](cognitive-search-working-with-skillsets.md).
 

articles/search/index.yml

@@ -12,7 +12,7 @@ metadata:
   ms.topic: landing-page
   author: HeidiSteen
   ms.author: heidist
-  ms.date: 09/04/2024
+  ms.date: 02/024/2025
 # linkListType: architecture | concept | deploy | download | get-started | how-to-guide | learn | overview | quickstart | reference | tutorial | video | whats-new
 
 landingContent:

articles/search/search-features-list.md

@@ -10,7 +10,7 @@ ms.service: azure-ai-search
 ms.custom:
   - ignite-2024
 ms.topic: conceptual
-ms.date: 09/04/2024
+ms.date: 02/24/2025
 ---
 
 # Features of Azure AI Search

articles/search/search-howto-index-sharepoint-online.md

@@ -9,7 +9,7 @@ ms.service: azure-ai-search
 ms.custom:
   - ignite-2023
 ms.topic: how-to
-ms.date: 08/20/2024
+ms.date: 02/24/2025
 ---
 
 # Index data from SharePoint document libraries

articles/search/search-indexer-field-mappings.md

@@ -11,7 +11,7 @@ ms.service: azure-ai-search
 ms.custom:
   - ignite-2023
 ms.topic: how-to
-ms.date: 08/09/2024
+ms.date: 02/24/2025
 ---
 
 # Field mappings and transformations using Azure AI Search indexers

articles/search/search-lucene-query-architecture.md

@@ -10,7 +10,7 @@ ms.service: azure-ai-search
 ms.custom:
   - ignite-2023
 ms.topic: conceptual
-ms.date: 08/19/2024
+ms.date: 02/24/2025
 ---
 
 # Full text search in Azure AI Search

articles/search/vector-search-filters.md

@@ -9,7 +9,7 @@ ms.service: azure-ai-search
 ms.custom:
   - ignite-2023
 ms.topic: how-to
-ms.date: 08/19/2024
+ms.date: 02/24/2025
 ---
 
 # Add a filter in a vector query in Azure AI Search
@@ -26,7 +26,7 @@ You can also use [Search Explorer](search-get-started-portal-import-vectors.md#c
 
 ## How filtering works in a vector query
 
-Filters apply to `filterable` nonvector fields, either a string field or numeric, to include or exclude search documents based on filter criteria. Although a vector field isn't filterable itself, filters can be applied to other fields in the same index, including or excluding the documents that also contain vector fields.
+Filters apply to `filterable` *nonvector* fields, either a string field or numeric, to include or exclude search documents based on filter criteria. Although a vector field isn't filterable itself, filters can be applied to other nonvector fields in the same index, including or excluding the documents that also happen to contain vector fields you're searching on.
 
 Filters are applied before or after query execution based on the `vectorFilterMode` parameter.
 

articles/search/vector-search-integrated-vectorization.md

@@ -1,38 +1,37 @@
 ---
 title: Integrated vectorization
 titleSuffix: Azure AI Search
-description: Add a data chunking and embedding step in an Azure AI Search skillset to vectorize content during indexing.
+description: Add a vector embedding step in an Azure AI Search skillset to vectorize content during indexing or queries.
 
 author: heidisteen
 ms.author: heidist
 ms.service: azure-ai-search
 ms.custom:
   - ignite-2023
 ms.topic: conceptual
-ms.date: 09/04/2024
+ms.date: 02/24/2025
 ---
 
-# Integrated data chunking and embedding in Azure AI Search
+# Integrated vector embedding in Azure AI Search
 
 Integrated vectorization is an extension of the indexing and query pipelines in Azure AI Search. It adds the following capabilities:
 
-+ Data chunking during indexing
-+ Text-to-vector conversion during indexing
-+ Text-to-vector conversion during queries
++ Vector encoding during indexing
++ Vector encoding during queries
 
-Data chunking isn't a hard requirement, but unless your raw documents are small, chunking is necessary for meeting the token input requirements of embedding models.
+[Data chunking](vector-search-how-to-chunk-documents.md) isn't a hard requirement, but unless your raw documents are small, chunking is necessary for meeting the token input requirements of embedding models.
 
-Vector conversions are one-way: text-to-vector. There's no vector-to-text conversion for queries or results (for example, you can't convert a vector result to a human-readable string).
+Vector conversions are one-way: nonvector-to-vector. For example, there's no vector-to-text conversion for queries or results, such as converting a vector result to a human-readable string, which is why indexes contain both vector and nonvector fields.
 
-Integrated data chunking and vectorization speeds up the development and minimizes maintenance tasks during data ingestion and query time because there are fewer external components to configure and manage. This capability is now generally available.
+Integrated vectorization speeds up the development and minimizes maintenance tasks during data ingestion and query time because there are fewer operations that you have to implement manually. This capability is now generally available.
 
 ## Using integrated vectorization during indexing
 
-For data chunking and text-to-vector conversions, you're taking a dependency on the following components:
+For integrated data chunking and vector conversions, you're taking a dependency on the following components:
 
-+ [An indexer](search-indexer-overview.md), which retrieves raw data from a [supported data source](search-indexer-overview.md#supported-data-sources) and serves as the pipeline engine.
++ [An indexer](search-indexer-overview.md), which retrieves raw data from a [supported data source](search-indexer-overview.md#supported-data-sources) and drives the pipeline engine.
 
-+ [A vector index](search-what-is-an-index.md) to receive the chunked and vectorized content.
++ [A search index](search-what-is-an-index.md) to receive the chunked and vectorized content.
 
 + [A skillset](cognitive-search-working-with-skillsets.md) configured for:
 
@@ -94,10 +93,6 @@ Data chunking (Text Split skill) is free and available on all Azure AI services
 
 + Combine vector and text fields for hybrid search, with or without semantic ranking. Integrated vectorization simplifies all of the [scenarios supported by vector search](vector-search-overview.md#what-scenarios-can-vector-search-support).
 
-## When to use integrated vectorization
-
-We recommend using the built-in vectorization support of Azure AI Foundry. If this approach doesn't meet your needs, you can create indexers and skillsets that invoke integrated vectorization using the programmatic interfaces of Azure AI Search.
-
 ## How to use integrated vectorization
 
 For query-only vectorization:
@@ -146,7 +141,7 @@ Here are some of the key benefits of the integrated vectorization:
 
 + Automate indexing end-to-end. When data changes in the source (such as in Azure Storage, Azure SQL, or Cosmos DB), the indexer can move those updates through the entire pipeline, from retrieval, to document cracking, through optional AI-enrichment, data chunking, vectorization, and indexing.
 
-+ Batching and retry logic is built in (non-configurable). Azure AI Search has internal retry policies for throttling errors that surface due to the Azure OpenAI endpoint maxing out on token quotas for the embedding model. We recommend putting the indexer on a schedule (for example, every 5 minutes) so the indexer can process any calls that were throttled by the Azure OpenAI endpoint despite of the retry policies.
++ Batching and retry logic is built in (non-configurable). Azure AI Search has internal retry policies for throttling errors that surface due to the Azure OpenAI endpoint maxing out on token quotas for the embedding model. We recommend putting the indexer on a schedule (for example, every 5 minutes) so the indexer can process any calls that are throttled by the Azure OpenAI endpoint despite of the retry policies.
 
 + Projecting chunked content to secondary indexes. Secondary indexes are created as you would any search index (a schema with fields and other constructs), but they're populated in tandem with a primary index by an indexer. Content from each source document flows to fields in primary and secondary indexes during the same indexing run. 
 

articles/search/vector-search-overview.md

@@ -9,7 +9,7 @@ ms.service: azure-ai-search
 ms.custom:
   - ignite-2023
 ms.topic: conceptual
-ms.date: 08/05/2024
+ms.date: 02/24/2025
 ---
 
 # Vectors in Azure AI Search
@@ -20,7 +20,7 @@ Vector search is an approach in information retrieval that supports indexing and
 + multilingual content ("dog" in English and "hund" in German)
 + multiple content types ("dog" in plain text and a photograph of a dog in an image file)
 
-This article provides [a high-level introduction to vectors](#vector-search-concepts) in Azure AI Search. It also explains integration with other Azure services and covers [terminology and concepts](#vector-search-concepts) related to vector search development.
+This article provides [a high-level introduction to vector support](#vector-search-concepts) in Azure AI Search. It also explains integration with other Azure services and covers [terminology and concepts](#vector-search-concepts) related to vector search development.
 
 We recommend this article for background, but if you'd rather get started, follow these steps:
 
@@ -39,9 +39,9 @@ Scenarios for vector search include:
 
 + **Search across different content types (multimodal)**. Encode images and text using multimodal embeddings (for example, with [OpenAI CLIP](https://github.com/openai/CLIP) or [GPT-4 Turbo with Vision](/azure/ai-services/openai/whats-new#gpt-4-turbo-with-vision-now-available) in Azure OpenAI) and query an embedding space composed of vectors from both content types.
 
-+ [**Hybrid search**](hybrid-search-overview.md). In Azure AI Search, hybrid search refers to vector and keyword query execution in the same request. Vector support is implemented at the field level, with an index containing both vector fields and searchable text fields. The queries execute in parallel and the results are merged into a single response. Optionally, add [semantic ranking](semantic-search-overview.md) for more accuracy with L2 reranking using the same language models that power Bing.
++ [**Hybrid search**](hybrid-search-overview.md). In Azure AI Search, we define hybrid search as dual vector and keyword query execution in the same request. Vector support is implemented at the field level. If an index contains both vector and non-vector fields, you can write a query that targets both. The queries execute in parallel and the results are merged into a single response and ranked accordingly.
 
-+ **Multilingual search**. Providing a search experience in the users own language is possible through embedding models and chat models trained in multiple languages. If you need more control over translation, you can supplement with the [multi-language capabilities](search-language-support.md) that Azure AI Search supports for nonvector content, in hybrid search scenarios.
++ **Multilingual search**. Azure AI Search is designed for extensibility. If you have embedding models and chat models trained in multiple languages, you can call them through custom or built-in skills on the indexing side, or vectorizers on the query side. If you need more control over text translation, you can supplement with the [multi-language capabilities](search-language-support.md) that Azure AI Search supports for nonvector content, in hybrid search scenarios.
 
 + **Filtered vector search**. A query request can include a vector query and a [filter expression](search-filters.md). Filters apply to text and numeric fields, and are useful for metadata filters, and including or excluding search results based on filter criteria. Although a vector field isn't filterable itself, you can set up a filterable text or numeric field. The search engine can process the filter before or after the vector query executes.
 
@@ -57,11 +57,11 @@ The following diagram shows the indexing and query workflows for vector search.
 
 On the indexing side, Azure AI Search takes vector embeddings and uses a [nearest neighbors algorithm](vector-search-ranking.md) to place similar vectors close together in an index. Internally, it creates vector indexes for each vector field.
 
-How you get embeddings from your source content into Azure AI Search depends on whether you want to perform the work within an Azure AI Search indexing pipeline, or externally.  Azure AI Search offers [integrated data chunking and vectorization](vector-search-integrated-vectorization.md) in an indexer pipeline. You still provide the resources (endpoints and connection information to Azure OpenAI), but Azure AI Search makes all of the calls and handles the transitions. This approach requires an indexer, a supported data source, and a skillset that drives chunking and embedding. Otherwise, you can handle all vectorization separately, and then push prevectorized content to [vector fields](vector-search-how-to-create-index.md) in a vector store.
+How you get embeddings from your source content into Azure AI Search depends on whether you want to perform the work within an Azure AI Search indexing pipeline, or externally.  Azure AI Search offers [integrated data chunking and vectorization](vector-search-integrated-vectorization.md) in an indexer pipeline. You still provide the resources (endpoints and connection information to Azure OpenAI), but Azure AI Search makes all of the calls and handles the transitions. This approach requires an indexer, a supported data source, and a skillset that drives chunking and embedding. If you don't want to use indexers, you can handle all vectorization externally, and then push prevectorized content into [vector fields](vector-search-how-to-create-index.md) in the search index.
 
 On the query side, in your client application, you collect the query input from a user, usually through a prompt workflow. You can then add an encoding step that converts the input into a vector, and then send the vector query to your index on Azure AI Search for a similarity search. As with indexing, you can deploy the [integrated vectorization](vector-search-integrated-vectorization.md) to convert the question into a vector. For either approach, Azure AI Search returns documents with the requested `k` nearest neighbors (kNN) in the results.
 
-Azure AI Search supports [hybrid scenarios](hybrid-search-overview.md) that run vector and keyword search in parallel, returning a unified result set that often provides better results than just vector or keyword search alone. For hybrid, vector and nonvector content is ingested into the same index, for queries that run side by side.
+Azure AI Search supports [hybrid scenarios](hybrid-search-overview.md) that run vector and keyword search in parallel, returning a unified result set that often provides better results than just vector or keyword search alone. For hybrid, vector and non-vector content is ingested into the same index, for queries that run side by side.
 
 ## Availability and pricing
 
@@ -88,7 +88,7 @@ Azure AI Search is deeply integrated across the Azure AI platform. The following
 | Azure AI Foundry | In the chat with your data playground, **Add your own data** uses Azure AI Search for grounding data and conversational search. This is the easiest and fastest approach for chatting with your data. |
 | Azure OpenAI | Azure OpenAI provides embedding models and chat models. Demos and samples target the [text-embedding-ada-002](/azure/ai-services/openai/concepts/models#embeddings-models). We recommend Azure OpenAI for generating embeddings for text. |
 | Azure AI Services | [Image Retrieval Vectorize Image API(Preview)](/azure/ai-services/computer-vision/how-to/image-retrieval#call-the-vectorize-image-api) supports vectorization of image content. We recommend this API for generating embeddings for images. |
-| Azure data platforms: Azure Blob Storage, Azure Cosmos DB | You can use [indexers](search-indexer-overview.md) to automate data ingestion, and then use [integrated vectorization](vector-search-integrated-vectorization.md) to generate embeddings. Azure AI Search can automatically index vector data from two data sources: [Azure blob indexers](search-howto-indexing-azure-blob-storage.md) and [Azure Cosmos DB for NoSQL indexers](search-howto-index-cosmosdb.md). For more information, see [Add vector fields to a search index.](vector-search-how-to-create-index.md). |
+| Azure data platforms: Azure Blob Storage, Azure Cosmos DB, Azure SQL, OneLake | You can use [indexers](search-indexer-overview.md) to automate data ingestion, and then use [integrated vectorization](vector-search-integrated-vectorization.md) to generate embeddings. Azure AI Search can automatically index vector data from [Azure blob indexers](search-howto-indexing-azure-blob-storage.md), [Azure Cosmos DB for NoSQL indexers](search-howto-index-cosmosdb.md), [Azure Data Lake Storage Gen2](search-howto-index-azure-data-lake-storage.md), [Azure Table Storage](search-howto-indexing-azure-tables.md), [Fabric OneLake](search-how-to-index-onelake-files.md). For more information, see [Add vector fields to a search index.](vector-search-how-to-create-index.md). |
 
 It's also commonly used in open-source frameworks like [LangChain](https://js.langchain.com/docs/integrations/vectorstores/azure_aisearch).
 
@@ -98,7 +98,7 @@ If you're new to vectors, this section explains some core concepts.
 
 ### About vector search
 
-Vector search is a method of information retrieval where documents and queries are represented as vectors instead of plain text. In vector search, machine learning models generate the vector representations of source inputs, which can be text, images, or other content. Having a mathematic representation of content provides a common basis for search scenarios. If everything is a vector, a query can find a match in vector space, even if the associated original content is in different media or language than the query.
+Vector search is a method of information retrieval where documents and queries are represented as vectors instead of plain text. In vector search, machine learning models generate the vector representations of source inputs, which can be text, images, or other content. Having a mathematic representation of content provides a common language for comparing disparate content. If everything is a vector, a query can find a match in vector space, even if the associated original content is in different media or language than the query.
 
 ### Why use vector search