Update Chain Reranker topic about the array and examples (#330)

* Update Chain Reranker topic about the array and examples

* Updates based on comments

* Updates

www/docs/api-reference/search-apis/reranking.md

@@ -20,11 +20,22 @@ accurate results.
 
 We currently provides the following rerankers: 
 
-* [**Multilingual Reranker v1**](/docs/learn/vectara-multi-lingual-reranker) (`type=customer_specific` and `reranker_id=rnk_272725719`) for accurate results across over 100 languages.
-* [**Maximal Marginal Relevance (MMR) Reranker**](/docs/learn/mmr-reranker) (`type=mmr`) for diversifying results while maintaining relevance.
-* [**User Defined Function Reranker**](/docs/learn/user-defined-function-reranker) (`type=userfn`) for custom scoring based on metadata.
-* [**Chain Reranker**](/docs/learn/chain-reranker) (`type=chain`) for combining multiple reranking strategies in sequence to meet 
-  more complex search requirements.
+* [**Multilingual Reranker v1**](/docs/learn/vectara-multi-lingual-reranker) (`type=customer_specific` and 
+  `reranker_name=Rerank_Multilingual_v1`) provides more accurate neural ranking than the 
+  initial Boomerang retrieval. While computationally more expensive, it offers 
+  improved text scoring across a wide range of languages, making it suitable 
+  for diverse content.
+* [**Maximal Marginal Relevance (MMR) Reranker**](/docs/learn/mmr-reranker) (`type=mmr`) for diversifying 
+  results while maintaining relevance.
+* [**User Defined Function Reranker**](/docs/learn/user-defined-function-reranker) (`type=userfn`) for custom 
+  scoring based on metadata.
+
+### Chain reranking
+
+The Vectara Chain Reranker (`type=chain`) lets you combine multiple reranking 
+strategies in sequence to meet more complex search requirements. This lets you 
+completely customize the functionality of Vectara to your needs by giving you 
+absolute control over the ranking functions. For details, see [Chain Reranker](/docs/learn/chain-reranker).
 
 ## Enable reranking
 
@@ -42,7 +53,7 @@ this simplified example intentionally omits several parameter values.
 {
   "query": "What is my question?",
   "stream_response": false,
-    "search": {
+  "search": {
     "start": 0,
     "limit": 25,
     "context_configuration": {},

www/docs/api-reference/search-apis/search.md

@@ -191,8 +191,11 @@ more rarely, as a response.
 The `reranker` object enables the reranking of query results, to further 
 increase relevance in certain scenarios. For more details, see [Rerank Search Results](/docs/api-reference/search-apis/reranking):
 
-* Specify the `type` as `customer_reranker` to use the [Multilingual Reranker v1](/docs/api-reference/search-apis/reranking#vectara-multilingual-reranker-v1).
-* Specify the `type` as `mmr` to use the [Maximal Marginal Relevance (MMR) Reranker](/docs/learn/mmr-reranker). This reranker lets you specify a `diversity_bias` value between `0.0` and `1.0`. 
+* Specify the `type` as `customer_reranker` and `reranker_name` as 
+  `Rerank_Multilingual_v1` to use the [Multilingual Reranker v1](/docs/api-reference/search-apis/reranking#vectara-multilingual-reranker-v1), 
+  also known as Slingshot.
+* Specify the `type` as `mmr` to use the [Maximal Marginal Relevance (MMR) Reranker](/docs/learn/mmr-reranker).
+  This reranker lets you specify a `diversity_bias` value between `0.0` and `1.0`. 
 * Specify the `type` as `userfn` to use the [User Defined Function Reranker(/docs/learn/user-defined-function-reranker)].
 * Specify the `type` as `chain` to use the [Chain Reranker](/docs/learn/chain-reranker).
 * If you do not want to use a reranker, set the type to `none`.

www/docs/learn/chain-reranker.md

@@ -9,18 +9,48 @@ import {vars} from '@site/static/variables.json';
 
 When refining query results where relevance and diversity are critical to 
 getting the best answers, relying on a single ranking strategy may not be 
-enough for business-critical search scenarios. The Chain Reranker addresses 
-this challenge by enabling you to apply two ranking strategies sequentially. 
+flexible enough for some scenarios. The Vectara Chain Reranker 
+addresses this challenge by enabling you to apply multiple ranking strategies 
+sequentially, giving you absolute control over the ranking process. 
 
-This is particularly useful in complex search scenarios that require multiple 
-dimensions of relevance. For example, enhancing diversity with the 
-[Maximal Marginal Relevance (MMR) reranker](/docs/learn/mmr-reranker) while also boosting documents 
-with specific metadata.
+This powerful capability enables you to completely customize Vectara's 
+functionality to your specific needs, ensuring that your search results and 
+generative AI applications align with your business requirements precisely.
+
+## Why use the chain reranker?
+
+Chaining multiple rerankers is particularly useful in complex search scenarios
+that require several dimensions of ranking. For example, enhancing 
+diversity with the [Maximal Marginal Relevance (MMR) reranker](/docs/learn/mmr-reranker), refining 
+the results across multiple languages, and then boosting documents with 
+specific metadata or other custom business logic with the 
+[User Defined Function reranker](/docs/learn/user-defined-function-reranker) to finalize the results.
+
+1. **Multidimensional relevance**: Combine multiple ranking strategies in sequence 
+   to optimize query results across various dimensions and meeting complex 
+   business scenarios.
+2. **Enhance RAG outcomes**: Improve the retrieval process by sequentially 
+   applying multiple rerankers that balance relevance, diversity, and custom 
+   business priorities to produce more contextually appropriate AI-generated 
+   responses in Retrieval Augmented Generation (RAG) systems.
+3. **Absolute control**: Tailor the ranking process to your exact specifications, 
+   ensuring that the query results reflect your custom business logic.
+
 
 ## Enable the chain reranker
 
-To enable the chain reranker, set the `type` to `chain` and then specify the 
-two rerankers that you want to combine.
+To enable the chain reranker, set the `reranker` object `type` to `chain` and then 
+specify the multiple rerankers that you want to combine in a `rerankers` array. 
+You can specify up to 50 rerankers in a chain.
+
+## Chain reranker examples
+
+In this example, the Vectara Multilingual Reranker first refines the results 
+with improved text scoring, while the [User Defined Function reranker](/docs/learn/user-defined-function-reranker) 
+then boosts relevance based on metadata for popularity. This simple chain 
+example can be useful for multilingual content platforms such as blogs, news 
+aggregators, and knowledge bases.
+
 
 ```json
 {
@@ -40,7 +70,53 @@ two rerankers that you want to combine.
 }
 
 ```
+Now let's take a look at a more complex chain reranker example. In this 
+e-commerce example, the Vectara Multilingual Reranker first refines the 
+initial results, providing a better ranking order than Boomerang. This 
+improves the overall relevance of the product results. The MMR reranker then 
+diversifies the results so that the customer sees some variety, rather than 
+only very similar products.
+
+```json
+{
+  "type": "mmr",
+  "diversity_bias": 0.2
+},
+```
+
+We set the diversity bias to `0.2` to have more emphasis on relevance while 
+still providing some diversity. You can increase this value to `0.3` or `0.4` 
+for more diversity.
+
+Finally, we apply a User Defined Function that combines several e-commerce 
+factors including a relevance score boost of 60%, a 20% boost if the product 
+is currently in stock, and then another 20% boost that favors products with 
+better customer reviews.
+
+This chain reranker maintains a balance between showing relevant products that 
+match what the customer is looking for, ensuring that the products are in 
+stock, and also takes into account customer satisfaction rating of the 
+products.
+
+```json
+{
+  "reranker": {
+    "type": "chain",
+    "rerankers": [
+      {
+        "type": "customer_specific",
+        "reranker_name": "Rerank_Multilingual_v1"
+      },
+      {
+        "type": "mmr",
+        "diversity_bias": 0.2
+      },
+      {
+        "type": "userfn",
+        "user_function": "get('$.metadata.relevance_score') * 0.6 + (get('$.metadata.in_stock') ? 0.2 : 0) + get('$.metadata.customer_rating') * 0.2"
+      }
+    ]
+  }
+}
+```
 
-In this example, the Vectara Multilingual Reranker refines the results across 
-over 100 languages, while the [User Defined Function reranker](/docs/learn/user-defined-function-reranker) boosts 
-relevance based on metadata for popularity.

www/docs/learn/vectara-multi-lingual-reranker.md

@@ -9,18 +9,45 @@ import TabItem from '@theme/TabItem';
 import {Config} from '@site/docs/definitions.md';
 import {vars} from '@site/static/variables.json';
 
-The new Vectara Multilingual Reranker V1 is a state-of-the-art reranking model 
-that significantly enhances the precision of retrieved results across 100+ 
-languages. Using this reranker requires both the `type` and `reranker_id` in the 
-`reranker` object. Set the `type` as `customer_reranker` and the `reranker_id` as `rnk_272725719`. 
+Generative AI applications often struggle with ranking the most relevant 
+information, leading to hallucinations and irrelevant responses. The new 
+Vectara Multilingual Reranker V1, also known as Slingshot, is a 
+state-of-the-art reranking model that significantly enhances the precision of 
+retrieved results. Providing advanced neural ranking, it refines the output of 
+initial models like [Boomerang](https://vectara.com/blog/introducing-boomerang-vectaras-new-and-improved-retrieval-model/), 
+offering even more accurate document scoring and response quality in Retrieval 
+Augmented Generation (RAG) pipelines.
+
+The Vectara Multilingual Reranker operates as a second-pass refinement tool, 
+building on Boomerang’s high-recall capabilities. While Boomerang quickly 
+retrieves a broad set of relevant documents, the Multilingual Reranker 
+delivers more precise results, ensuring that the top-ranked documents are the 
+most relevant. This reranker also excels across both English and multilingual 
+datasets, making it a powerful tool for global use cases.
+
+While more computationally expensive and introducing some additional latency, 
+the multilingual reranker improves neural ranking beyond Boomerang’s initial 
+selection by providing more precise text scoring. Think of the Slingshot 
+reranker as a "better Boomerang" for refining results, with the multilingual 
+capability serving primarily as a differentiator from other rerankers in the 
+market, which are often English-only.
+
+Using this reranker requires both the `type` and `reranker_id` in the 
+`reranker` object. Set the `type` as `customer_reranker` and the `reranker_name` 
+as `Rerank_Multilingual_v1`. 
 
 ```json
 "reranker": {
       "type": "customer_reranker",
-      "reranker_id": "rnk_272725719"
+      "reranker_name": "Rerank_Multilingual_v1"
     }
 ```
 
+:::note
+The `reranker_id` and `rnk_272725719` have been deprecated. Use `reranker_name` and 
+`Rerank_Multilingual_v1` instead.
+:::
+
 The Vectara Multilingual Reranker ensures impressive zero-shot performance on 
 unseen data and domains, and it **never** trains on customer data. In RAG use 
 cases, this reranker distinguishes the scores of relevant and irrelevant