Cohere Rerank 3.5 is now available on Amazon Bedrock via Rerank API

We are pleased to announce the availability of Cohere's advanced reranking model, Rerank 3.5, through our new Rerank API on amazon Bedrock. This powerful reranking model enables AWS customers to significantly improve their search relevance and content ranking capabilities. This model is also available to users of the amazon Bedrock knowledge base. By incorporating Cohere's Rerank 3.5 into amazon Bedrock, we are making enterprise-grade search technology more accessible and empowering organizations to improve their information retrieval systems with minimal infrastructure management.

In this post, we discuss the need for Reranking, the capabilities of Cohere's Rerank 3.5, and how to get started with it on amazon Bedrock.

Reclassification for advanced recovery

Reclassification is a vital enhancement to retrieval augmented generation (RAG) systems that adds a sophisticated second layer of analysis to improve the relevance of search results beyond what traditional vector search can achieve. Unlike integration models that rely on pre-computed static vectors, reclassifiers perform dynamic query-time analysis of document relevance, allowing for more nuanced and contextual matching. This capability allows RAG systems to effectively balance between broad document retrieval and precise context selection, ultimately leading to more accurate and reliable results from language models while reducing the likelihood of hallucinations.

Existing search systems benefit significantly from reclassification technology by providing more contextually relevant results that directly impact user satisfaction and business results. Unlike traditional keyword matching or basic vector search, reranking performs an intelligent second-pass analysis that considers multiple factors, including semantic meaning, user intent, and business rules to optimize the order of search results. search. Specifically in ecommerce, reranking helps surface the most relevant products by understanding the nuanced relationships between search queries and product attributes, while incorporating crucial business metrics like conversion rates and inventory levels. This advanced relevance optimization leads to better product discovery, higher conversion rates, and higher customer satisfaction on digital commerce platforms, making reranking an essential component for any modern enterprise search infrastructure.

Introducing Cohere Rerank 3.5

Rerank 3.5 from Cohere is designed to improve search and RAG systems. This intelligent cross-coding model takes a query and a list of potentially relevant documents as input and then returns the documents sorted by semantic similarity to the query. Cohere Rerank 3.5 excels at understanding complex information that requires reasoning and is able to understand the meaning behind business data and user questions. Its ability to understand and analyze business data and user questions in more than 100 languages, including Arabic, Chinese, English, French, German, Hindi, Japanese, Korean, Portuguese, Russian and Spanish, makes it particularly valuable to global organizations in sectors such as finance, healthcare, hospitality, energy, government and manufacturing.

One of the key advantages of Cohere Rerank 3.5 is its ease of implementation. Through a single call to the Rerank API in amazon Bedrock, you can integrate Rerank into existing systems at scale, whether semantic or keyword-based. Reclassification strictly improves first-stage retrievals on standard text retrieval benchmarks.

Cohere Rerank 3.5 is the latest in the financial field, as illustrated in the figure below.

Cohere Rerank 3.5 is also the latest in the e-commerce space, as illustrated in the figure below. Cohere's e-commerce benchmarks revolve around the recovery of various products, including fashion, electronics, food and more.

The products were structured as strings in a key-value pair format like the following:

“Title”: “Title” 
“Description”: “Long-form description” “Type”:  etc.....

Cohere Rerank 3.5 also excels in hospitality, as shown in the figure below. Hospitality benchmarks revolve around recovering hospitality experiences and accommodation options.

The documents were structured as strings in a key-value pair format like the following:

“Listing Title”: “Rental unit in Toronto” “Location”: “171 John Street, Toronto, Ontario, Canada”

“Description”: “Escape to our serene villa with stunning downtown views....”

We see notable improvements in project management performance across all types of issue tracking tasks, as illustrated in the figure below.

Cohere project management benchmarks cover a variety of recovery tasks, such as:

• Search engineering tickets from various project management and issue tracking software tools.
• Search for GitHub issues in popular open source repositories

Get started with Cohere Rerank 3.5

To get started using Cohere Rerank 3.5 with Rerank API and amazon Bedrock Knowledge Bases, navigate to the amazon Bedrock console and click Access to the model in the left panel. Click on Modify accessselect Cohere Rerank 3.5, click Next and press submit.

Get started with the amazon Bedrock rerank API

The Cohere Rerank 3.5 model, powered by the amazon Bedrock Rerank API, allows you to reclassify input documents directly based on their semantic relevance to a user's query, without the need for a preconfigured knowledge base. The flexibility makes it a powerful tool for various use cases.

To get started, configure your environment by importing the necessary libraries and initializing the Boto3 clients:

import boto3
import json
region = boto3.Session().region_name

bedrock_agent_runtime = boto3.client('bedrock-agent-runtime',region_name=region)

modelId = "cohere.rerank-v3-5:0"
model_package_arn = f"arn:aws:bedrock:{region}::foundation-model/{modelId}”

Next, define a main function that reorders a list of text documents by calculating relevance scores based on the user's query:

def rerank_text(text_query, text_sources, num_results, model_package_arn):
    response = bedrock_agent_runtime.rerank(
        queries=(
            {
                "type": "TEXT",
                "textQuery": {
                    "text": text_query
                }
            }
        ),
        sources=text_sources,
        rerankingConfiguration={
            "type": "BEDROCK_RERANKING_MODEL",
            "bedrockRerankingConfiguration": {
                "numberOfResults": num_results,
                "modelConfiguration": {
                    "modelArn": model_package_arn,
                }
            }
        }
    )
    return response('results')

For example, imagine a scenario where you need to identify emails related to item returns from a multilingual data set. The following example demonstrates this process:

example_query = "What emails have been about returning items?"

documents = (
    "Hola, llevo una hora intentando acceder a mi cuenta y sigue diciendo que mi contraseña es incorrecta. ¿Puede ayudarme, por favor?",
    "Hi, I recently purchased a product from your website but I never received a confirmation email. Can you please look into this for me?",
    "مرحبًا، لدي سؤال حول سياسة إرجاع هذا المنتج. لقد اشتريته قبل بضعة أسابيع وهو معيب",
    "Good morning, I have been trying to reach your customer support team for the past week but I keep getting a busy signal. Can you please help me?",
    "Hallo, ich habe eine Frage zu meiner letzten Bestellung. Ich habe den falschen Artikel erhalten und muss ihn zurückschicken.",
    "Hello, I have been trying to reach your customer support team for the past hour but I keep getting a busy signal. Can you please help me?",
    "Hi, I have a question about the return policy for this product. I purchased it a few weeks ago and it is defective.",
    "早上好，关于我最近的订单，我有一个问题。我收到了错误的商品",
    "Hello, I have a question about the return policy for this product. I purchased it a few weeks ago and it is defective."
)

Now, prepare the list of text fonts to be passed to the rerank_text() function:

text_sources = ()
for text in documents:
    text_sources.append({
        "type": "INLINE",
        "inlineDocumentSource": {
            "type": "TEXT",
            "textDocument": {
                "text": text,
            }
        }
    })

Then you can invoke rerank_text() specifying the user's query, text resources, desired number of top-ranked results, and model ARN:

response = rerank_text(example_query, text_sources, 3, model_package_arn)
print(response)

The output generated by the amazon Bedrock Rerank API with Cohere Rerank 3.5 for this query is:

({'index': 4, 'relevanceScore': 0.1122397780418396},
 {'index': 8, 'relevanceScore': 0.07777658104896545},
 {'index': 2, 'relevanceScore': 0.0770234540104866})

The relevance scores provided by the API are normalized to a range of (0, 1), with higher scores indicating greater relevance to the query. Here the 5^th The document list item is the most relevant. (Translated from German to English: Hello, I have a question about my last order. I received the wrong item and I need to return it.)

You can also get started using Cohere Rerank 3.5 with amazon Bedrock knowledge bases by completing the following steps:

1. In the amazon Bedrock console, choose Knowledge bases low Construction tools in the navigation panel.
2. Choose Create knowledge base.
3. Provide your knowledge base details such as name, permissions, and data source.
   

1. To configure your data source, specify the location of your data.
2. Select an embedding model to convert data to vector embeddings, and have amazon Bedrock create a vector store in your account to store the vector data.

When you select this option (available only in the amazon Bedrock console), amazon Bedrock creates a vector index on amazon OpenSearch Serverless (by default) in your account, eliminating the need to manage anything yourself.

1. Review your settings and create your knowledge base.
2. In the amazon Bedrock console, choose your knowledge base and choose Test knowledge base.
   
3. Choose the icon for additional configuration options to test your knowledge base.
4. Choose your model (for this post, Cohere Rerank 3.5) and choose Apply.

The settings panel displays the new Reclassification Section menu with additional configuration options. The number of source fragments reclassified returns the specified number of highest relevant fragments.

Conclusion

In this post, we explore how to use Cohere's Rerank 3.5 model on amazon Bedrock, demonstrating its powerful capabilities for improving search relevance and robust reranking capabilities for enterprise applications, improving user experience and optimizing workflows. information retrieval. Start improving your search relevance today with Cohere's Rerank model on amazon Bedrock.

Cohere Rerank 3.5 on amazon Bedrock is available in the following AWS regions: us-west-2 (US West – Oregon), ca-central-1 (Canada – Central), eu-central-1 (Europe – Frankfurt) and ap-northeast-1 (Asia Pacific – Tokyo).

Share your comments with <a target="_blank" href="https://repost.aws/tags/TAQeKlaPaNRQ2tWB6P7KrMag/amazon-bedrock” target=”_blank” rel=”noopener”>AWS re: Publishing for amazon Bedrock or through your usual AWS Support contacts.

To learn more about the features and capabilities of Cohere Rerank 3.5, see the Cohere product page on amazon Bedrock.

About the authors

Karan Singh is a generative ai specialist for third-party models at AWS, where he works with top-tier third-party core model (FM) providers to develop and execute joint go-to-market strategies, enabling customers to effectively train, deploy, and execute. . scale FM to solve industry-specific challenges. Karan holds a Bachelor of Science in Electrical and Instrumentation Engineering from Manipal University, a Master of Science in Electrical Engineering from Northwestern University, and is currently an MBA candidate at the Haas School of Business at the University of California, Berkeley.

James Yi is a Senior ai/ML Partner Solutions Architect at amazon Web Services. He leads AWS strategic partnerships in emerging technologies, guiding engineering teams to design and develop joint cutting-edge solutions in generative ai. It enables technical and field teams to seamlessly deploy, operate, secure, and integrate partner solutions on AWS. James collaborates closely with business leaders to define and execute joint go-to-market strategies, driving cloud-based business growth. Outside of work, he enjoys playing soccer, traveling, and spending time with his family.