Empowering Virtual Assistants: 8 Key Strategies for Effectively Training LLMs
Jenny Machado August 31, 2023
In the rapidly evolving technology landscape, large Language Models (LLMs) have emerged as a leading innovation, transforming the way we interact with technology and opening doors to new possibilities. Training Deep Learning-based Language Models to have competent conversations with artificial intelligence is an exciting challenge that involves a combination of supervised learning techniques, reinforcement, and contextual adaptation. Virtual Assistants powered by Deep Learning-based Language Models (LLM) have transformed the way we interact with technology. In this article, we will explore 8 key strategies for training LLMs and building virtual assistants capable of delivering exceptional experiences.

8 Key Strategies for Effectively Training LLMs

  1. Conversation Data Training
The first step in improving an LLM's conversational skills is to provide him or her with a large amount of conversational data. This includes dialogues and discussions covering a variety of topics and interaction styles. This data should cover a range of topics and situations so that the assistant can understand and respond to a wide range of queries.
  1. Reinforcement Learning
Supervised training is essential for the attendant to learn to generate consistent and accurate responses. However, combining it with reinforcement learning can take it to a higher level. Through user feedback and evaluation of the quality of responses, the assistant can learn to improve its responses based on experience.
  1. Advanced Context Modeling
Understanding context is critical in a meaningful conversation. LLMs must be trained to understand not only the current query, but also the context of the previous conversation. This ensures that responses are relevant and consistent throughout the interaction.
  1. Real-time user feedback
User feedback is a valuable source of improvement for virtual assistants. Providing an easy way for users to rate and give feedback on the assistant's responses can help to continuously adjust the model and improve its capabilities. These evaluations can be used as feedback signals to refine the model during the training process.
  1. Generating creative responses
A powerful virtual assistant not only provides accurate, but also creative and natural responses. LLMs must be trained to generate responses that do not sound robotic but reflect how a human being might respond in a similar situation. Models must learn to avoid offensive, inappropriate or misleading responses.
  1. Adaptation to Individual Users
Users have unique conversational styles and preferences. Advanced LLMs can be trained to adapt to a specific user over time. This can be accomplished by allowing the model to interact with the same user multiple times and learn from their choices and feedback.
  1. Multilingual and Cultural Integration
A powerful virtual assistant must be able to understand and respond in multiple languages and cultural contexts. Training in multilingual data and diverse cultural expressions is essential to achieve this capability.
  1. Rigorous Testing and Continuous Optimization
Once the virtual assistant has been implemented, rigorous testing is crucial. Potential problems must be identified and addressed, erroneous responses corrected, and the model adjusted based on actual usage results.

eva: orchestrating LLMs

LLMs have proven to be a milestone in AI and their journey promises exciting and transformative terrain. Their future is characterized by constant improvement, increased personalization, real-world applications, and a pivotal role in creativity and education. As these models become more conversationally proficient, they will be able to play a more integral role in a variety of applications, from advanced virtual assistants to customer support systems and beyond. At NTT DATA, we go together with technological innovation, which is why our eva platform has created an orchestrator for LLMs that simplifies complex interactions. This new functionality enhances our platform's ability to orchestrate calls to orchestrate calls to generative AI tools, such as Azure OpenAI services, making it easier to handle more advanced and complex tasks with unprecedented simplicity and elegance. At eva, we use a variety of generative AI models provided by Azure OpenAI (and other vendors) to meet various needs, such as content generation, classification, and data processing.
Must News
NTT DATA partners with L’ Oréal to enhance its digital e-commerce platforms
Jenny Machado April 24, 2023

NTT DATA partners with L'Oréal to enhance its digital e-commerce platforms

L'Oréal's goal is to provide tools that are at the forefront of consumer experience technology, delivering added value using artificial intelligence. NTT DATA announced the recent alliance with the leading global cosmetics and beauty company, L'Oréal Groupe, which seeks to develop innovative technological solutions that allow users a personalized experience at any time of the day they need it through digital channels, also improving customer service responses. Since its inception, L'Oréal Groupe has been characterized as a company that constantly embraces innovation, and this is no exception. As a pioneer in beauty-tech, its goal is to push the boundaries of beauty, using science and technology as the basis of its sustainability strategy, satisfying the desires and dreams of all its consumers around the world. On this occasion, both companies indicated that they are working on technologies that will be pioneers in Latin America, and are being developed through NTT DATA's Conversational AI, eva platform, which uses advanced artificial intelligence with Generative AI, capable of understanding and offering hyper-personalized advice to users. "As a company we are constantly researching and trying to develop processes that facilitate and optimize the shopping experience for our consumers. We are excited to bring cutting-edge technology to our customers, we want to explore the use of tools like GenAI to provide an engaging and personalized shopping experience, helping our customers find the best products for their beauty needs," said Arturo Perez Wong Manager Manager DPGP L'Oréal Chile. At the same time, the companies added that this alliance will allow them to capitalize on the advancement of the latest artificial intelligence and big data technologies and apply them in the multiple communication channels they have with their customers, creating a more dynamic, fluid, and intuitive shopping experience in both online and offline modalities. "We are very happy to be able to contribute with all our potential to accelerate the deployment of Social Commerce at L'Oréal, and help them to improve the customer experience, through a new, simpler and revolutionary way of shopping, which saves time and improves users' lives" concluded Santiago Santa María, Director Conversational AI & Generative AI.  

About L'Oréal

L'Oréal, a world leader in beauty and personal care, is headquartered in Paris, France. With more than a century of history, the company operates in 150 countries and employs around 88,000 employees. L'Oréal's diverse portfolio includes brands such as L'Oréal Paris, NYX, Vogue and Maybelline. The company focuses on innovation, sustainability, and social responsibility, offering high-quality products and personalized beauty experiences for its customers around the world.  


NTT DATA, part of the NTT Group, is an innovative global IT and business services company headquartered in Tokyo. The company helps clients in their transformation process through consulting, industry solutions, business process services, digital and IT modernization and managed services. NTT DATA enables them, as well as society, to move confidently into the digital future. The company demonstrates its commitment to the long-term success of its customers by combining global reach with local focus to work with them in more than 50 countries around the world.
Must News
GenAI: the essential tool for digital customer service in today’s business world
Jenny Machado March 30, 2023
GenAi has turned virtual conversations around with its powerful capabilities, it is amazing how it manages to understand the context and intentions behind thousands of questions to answer users through natural language in an accurate way. This is on everyone's lips; some companies are already riding the wave with this technology developed by OpenAI. A tool trained to perform different tasks related to natural language has become a great plus for the creation of virtual assistants. The application manages to generate texts in a coherent and natural way, improves accuracy in information search systems, enhances the development of chatbots by responding to users in a concise manner, and can also be used to improve natural language processing.

But why is GenAI so attractive for businesses? We dared to ask the tool itself and this is a summary of its answer:

It improves customer service, reduces costs, analyzes data about your customer interactions, responds quickly to questions or queries, can handle multiple conversations, collects customer information to provide personalized communication. Also, we sought the voices of experts to delve into the topic, Santiago Santa Maria, director of Conversational AI and ChatGPT at NTT DATA, talks about the revolution in customer care and how GenAI is changing the game. In his article published on Medium, Santiago Santa Maria, points out that customer care is being radically transformed with this new application.  
"We are still shocked by the disruption that GenAI is causing in multiple industries: content generation (texts, books, scripts articles...), summaries, translations, answers to factual questions... New uses and applications of this powerful technology keep appearing" says the director of NTT DATA.
  He explains how a company can benefit from GenAI to improve customer service, as it is possible to connect this Extensive Language Model (Generative AI ) with different digital channels such as WhatsApp, Instagram, Web, and Call Centers.

Other benefits identified by the director of Conversational AI and GenAI are:

  • Understanding and accurately responding to user needs.
  • Providing solutions to complex user problems
  • Deliver personalized and satisfying user experiences
  • Handle multiple languages and dialects
  • Understanding context and providing relevant responses across different digital channels
  • Deliver personalized recommendations and suggestions to users
  • Handle high emotional stress situations and provide empathy and appropriate solutions
  • Increase efficiency and reduce costs associated with customer support and sales.
He also gives us a view from his experience, of the value of GenAI to improve sales by connecting it to digital channels. His vision regarding Contact Centers is that GenAI has the potential to automate tasks that today are performed by many people. With this application, cases would be automated to free up hours of attention by Contact Center executives.
"GenAi has shown us the disruptive change that Generative AI bring to the Conversational AI industry."
NTT DATA’s Conversational AI platform, eva, already makes uses Generative AI and is powered by OpenAI technology to complement its capabilities and help clients boost their customer service processes, drive up sales, reduce costs, increase efficiency, and improve their users' experience, giving brands a more salient vision.
Must News
NTT DATA has been recognized as a Major Contender in Everest Group’s 2022 Conversational AI PEAK Matrix® Assessment
Jenny Machado November 15, 2022
We are pleased to announce that NTT DATA has been recognized as a Major Contender in Everest Group’s 2022 Conversational AI PEAK Matrix® Assessment This year’s PEAK Matrix® Assessment evaluates 26 global Conversational AI vendors. Everest Group have analyzed market impact, vision and capability to deliver services. NTT DATA ranks as one of the strongest vendors in market impact and one of the top 10 Conversational AI vendors.  The report provides an objective, data-driven comparative assessment of service and technology providers based on their overall capability and market success. In this quadrant, Everest Group has evaluated competitors in the market, naming NTT DATA as one of the main competitors thanks to its Conversational AI solution. From Everest Group they highlighted the strengths of eva, among them it is highlighted that the platform is made by a multidisciplinary team of professionals (technical architects, development specialists, data engineers, linguists, UX writers and QA teams); In addition, the platform is aimed at business users; Lastly, it offers native support in 53 languages ​​and has multiple developments in Portuguese, Spanish, English and Italian.  eva, NTT DATA's enterprise conversational AI platform for creating and managing an unlimited number of virtual agents, helps contact centers handle written or spoken conversations accurately and at scale, while reducing the cost of service and improving the user experience. With its powerful, no-code Dialog Manager, it enables conversational designers to create automated transactional conversations across any channel. eva NLP delivers multilingual cognition and automated learning using a modern open architecture that enables the scalability and transactional power demanded by large organizations. This demonstrates the efforts of a great team, who are focused on developing a high-level product that is helping large companies through its capabilities, constant improvements and artificial intelligence. We are proud to appear in the Everest Group quadrant. View the Everest Group quadrant here  
Must News
Say hello to the great new eva features, to create virtual assistants
Jenny Machado September 29, 2022

Discover what's new for creating virtual assistants

We are bringing to you new possibilities and improvements to our NTT DATA eva platform that will bring your experience when creating virtual assistants to a whole new level.
Check out the new features eva by NTT DATA has been cooking up for you.


You have asked for it, you shall have it! We are glad to announce that we will now have dashboard with KPIs that will help you measure your Virtual Agent's performance and keep improving it. With a magical visualization, you will be able to analyze if you’re achieving your business goals. Total conversations Number of sessions between the virtual agent and the users on a given period and its percentage change compared to the previous period. Total conversations Number of sessions on a time frame. Total messages Total number of messages sent by users during conversations and its percentage change compared to the previous period. Total messages Number of user messages on a time frame. % accuracy Model’s hit ratio: it calculates the number of user messages that did not end in a Not Expected cell divided by the total of user messages and its percentage change compared to the previous period. Total of users Number of users (new and returning) who started a new session and its percentage change compared to the previous period. This information requires that the business key is informed. Total of users Variation of users (new and returning) who started a new session on a given period. This information requires that the business key is informed. Top 10 intents The 10 most accessed intents (as returned by the NLP) and their occurrences by channel. Top 10 flows The 10 most executed user journey flows


We have added an improvement on the way navigate in the repositories. Control how many items are displayed on the repository lists. Choose if you want to see 5 or 100 items per page on the flows, intents, entities, services, and answers repositories.


Have you ever wanted to organize the repository by its name, modification date, or its type? Now you can do it! We added this new feature in all repositories to organize them as you want and to improve the search of all created items.


Another improvement that will ease your experience and save time! Choose how you want to import your virtual agent: as a new virtual agent or update a current version (parameters, channels, workspace, repositories, and Automated Learning).  This option will allow you move the virtual agent through environments directly from the virtual agent popup menu, without the need of creating a new one every time a change is made in other environment.


We update Automated Learning. Do you know this feature? Automated Learning allows training virtual agent from documents, turning the trainings into an easy task. We have added improvements such as the possibility to add questions to disabled documents. Now, get ready to take off by creating virtual assistants, with a coffee in your hand and your best smile  We are part of NTT DATA's Syntphony ecosystem
Must News
NTT Data and Voximplant: A powerful alliance that will revolutionize call centers
Jenny Machado February 22, 2022
The alliance of NTT Data and Voximplant announce eva Voice Cloud, a powerful technological solution that offers digital transformation in call centers to companies, with a 100% cloud solution that revolutionizes user experience. NTT DATA, a leading company in transformation, technology and consulting operations in Europe, the United States and LATAM, strengthens its alliance with Voximplant, a leader in cloud call center solutions. The partnership to revolutionize call centers with Conversational AI will be through eva, the platform to create and manage virtual agents created by NTT Data experts which now extends its capabilities in the call center with Voximplant. Together they will offer the best cloud solution to transform the customer experience in call centers. Companies understand they must ensure a high level of user experience, and need to react robustly to new market conditions. User satisfaction when interacting with call centers is very low. Most companies fail to offer a call center service that meets the customer's expectation. Strategic alliances such as  NTT DATA and Voximplant, aim to shed light on this need with an agile and efficient solution. "The most important thing for us is agility, being able to deploy call center solutions in short times where we automate phone calls," says Santiago Santa María Director of Conversational AI at NTT DATA. We created eva Voice Cloud powered by Voximplant, a robust and  complete business solution, which uses artificial intelligence to improve the user experience in voice conversations. "Collaboration is born from the need to drive improvement in care; the pandemic has forced companies to make courageous decisions. Today, migrating to the cloud is a necessity that accelerates the transformation of customer service," says Santiago. No doubt remote work and social distancing increased call center calls.  eva by NTT Data provides robustness and quality in the cloud, thus allowing a faster dialogue with customers. This alliance is ambitious and promising, as both companies have a wide commercial expectation in the use of AI in call centers. Responding to the user immediately and solving their need the first time is a priority and a fundamental requirement to compete in the market of today and the future.
Must News
Agent Template: an easy way to create your virtual assistant 
Jenny Machado January 19, 2022
Creating virtual agents with eva is now faster and easier with the new templates we have designed for you. What can take up to two months of development between research, writing, flow creation, testing and more, can now be reduced to less than a week with Agent Template's.
Agent Template is a collection of flows provided by eva that can be used to establish a base for building conversations.

Currently available:

  • Banking Agent Template, with 37 ready-made flows and 5 use cases for financial services. Available in english, spanish and portuguese.
  • Foundation Agent Template, with 13 flows common to many industries and sectors, such as NPS, Welcome, and Talk to an Agent. Available in English, Spanish and Portuguese.
  • Healthcare Agent Template is a collection of 18 flows focused on healthcare service that can be used to establish a base for building conversations. Available in English, Spanish and Portuguese.
  • Ticketing Agent Template is a collection of 19 flows focused on Ticketing service that can be used to establish a base for building conversations. Available in Spanish and Portuguese.
  • Telecom Agent Template is focused for Telecom virtual assistant, featuring 25 flows. Available in Spanish and Portuguese.
They were designed based on market best practices, with the goal of optimizing both your team and the process of building a virtual agent. We worked together with people from NTT DATA and Digital Experience. 

In agent templates, you will save time on your project because it offers:

  • Ready editable flows. In other words, your team won't have to worry about building the bot flows. You only need to edit the steps and adjust the responses to your business needs.
  • A bot ready for each industry, with specific use cases.
  • Text and walkthrough suggestions, material developed with UX best practices and various studies (such as benchmarks, interviews and user testing).
The consumer will only have to customize the information and webhooks according to their business case.    We have created three versions: Spanish, Portuguese and English, and a manual to help your team.    Agent Templates can speed up many tasks and allow teams to act more strategically. New agent templates for different industries and use cases will be added soon You can download the agent template on Github You may be interested in reading this article: 4 reasons for banks to bet on AI Conversational
Must News
Masterclass on Conversational AI
Jenny Machado September 23, 2021
Santiago Santa María, Director of conversational AI, participated in the Marketing 2021 Cycle organized by the Chilean Institute of Rational Business Administration ICARE to talk about Conversational AI. In the masterclass, Santiago explained how companies in different sectors have been able to boost their productivity, increase sales, reduce costs, and improve customer experience, thanks to automated conversations created with artificial intelligence. "Today, technological advances and AI make it possible to establish a good connection between companies and customers, and in addition, the benefits of automating processes include reducing costs, increasing sales, scaling and providing the best user experience" expressed Santiago. He also talked about the negative consequences of not creating a good virtual assistant, such as generating a bad experience for users and disconnecting them from the company or brand. Another important point he made was the transition in purchases in new sales channels, such as Instagram or WhatsApp. He indicates that in that sense, Conversational AI, allows scanning data to understand the behavior patterns of users or analyze their habits to guide customers through a specific message, achieving a process of communication and positive relationship between the consumer and the company.
"This allows us in an automated way to complete the entire purchase process and even make the payment in these channels. We are using virtual assistants to increase sales, they are becoming tireless salespeople, who have the capacity to attend 24/7 to as many people as necessary."
Would you like to listen to this masterclass where you will learn tips and recommendations to make a virtual assistant integrate and adapt to the brand values? Watch the Masterclass [embed]https://vimeo.com/616446556[/embed] We invite you to learn more about eva to create virtual assistants that adapt to the brand values and achieve a good experience.
Must News