はじめに
対話型AIプラットフォーム
チャットボットの概要
自然言語処理(NLP)
ボットの概念と用語
クイックスタートガイド
プラットフォームへのアクセス
ボットビルダーの操作
リリースノート
最新バージョン(英語)
以前のバージョン(英語)
コンセプト
設計
ストーリーボード
ダイアログタスク
ダイアログタスクとは
ダイアログビルダー
ノードタイプ
インテントノード
ダイアログノード
エンティティノード
フォームノード
確認ノード
ロジックノード
ボットアクションノード
サービスノード
Webhookノード
スクリプトノード
グループノード
エージェント転送ノード
ユーザープロンプト
音声通話プロパティ
イベント ハンドラー
ナレッジグラフ
ナレッジグラフの抽出
ナレッジグラフの構築
ボットにナレッジグラフを追加
グラフの作成
ナレッジグラフの構築
既存のソースからFAQを構築
通知タスク
スモールトーク
デジタルスキル
デジタルフォーム
デジタルビュー
デジタルビューとは
パネル
ウィジェット
トレーニング
トレーニングとは
機械学習
機械学習とは
モデル検証
ファンダメンタルミーニング
ナレッジグラフ
示唆
ランキングおよび解決
NLPの詳細設定
NLPのガイドライン
インテリジェンス
インテリジェンスとは
コンテキスト
コンテキストインテント
割り込み
複数インテントの検出
エンティティの変更
デフォルトの会話
センチメント管理
トーン分析
テストとデバッグ
ボットと会話
発話テスト
バッチテスト
会話テスト
デプロイ
チャネル
公開
LLM and Generative AI
Introduction
LLM Integration
Kore.ai XO GPT Module
Prompts & Requests Library
Co-Pilot Features
Dynamic Conversations Features
分析
ボットの分析
NLPメトリクス
会話フロー
Usage Metrics
封じ込め測定
カスタムダッシュボード
カスタムダッシュボードとは
メタタグ
カスタムダッシュボードとウィジェット
ユニバーサルボット
ユニバーサルボットとは
ユニバーサルボットの定義
ユニバーサルボットの作成
ユニバーサルボットのトレーニング
ユニバーサルボットのカスタマイズ
他言語の有効化
ストア
プラントと使用
Overview
Usage Plans
Support Plans
Invoices
管理
ボット認証
複数言語対応ボット
個人を特定できる情報の編集
ボット変数の使用
IVRのシステム連携
一般設定
ボット管理
ハウツー
Integrations
Actions
Actions Overview
Asana
Configure
Templates
Azure OpenAI
Configure
Templates
BambooHR
Configure
Templates
Bitly
Configure
Templates
Confluence
Configure
Templates
DHL
Configure
Templates
Freshdesk
Configure
Templates
Freshservice
Configure
Templates
Google Maps
Configure
Templates
Here
Configure
Templates
HubSpot
Configure
Templates
JIRA
Configure
Templates
Microsoft Graph
Configure
Templates
Open AI
Configure
Templates
Salesforce
Configure
Templates
ServiceNow
Configure
Templates
Stripe
Configure
Templates
Shopify
Configure
Templates
Twilio
Configure
Templates
Zendesk
Configure
Templates
Agents
Agent Transfer Overview
Custom (BotKit)
Drift
Genesys
Intercom
NiceInContact
NiceInContact(User Hub)
Salesforce
ServiceNow
Configure Tokyo and Lower versions
Configure Utah and Higher versions
Unblu
External NLU Adapters
Overview
Dialogflow Engine
Test and Debug
会話スキルの設計
バンキングボットを作成
バンキングボット – 資金の振り替え
バンキングボット – 残高を更新
ナレッジグラフを構築
スマートアラートの予約方法
デジタルスキルの設計
デジタルフォームの設定方法
デジタルビューの設定方法
データテーブルのデータの追加方法
データテーブルのデータの更新方法
Add Data from Digital Forms
ボットのトレーニング
示唆の使用方法
インテントとエンティティのパターンの使用方法
コンテキスト切り替えの管理方法
ボットのデプロイ
エージェント転送の設定方法
ボット関数の使用方法
コンテンツ変数の使用方法
グローバル変数の使用方法
ボットの分析
カスタムダッシュボードの作成方法
カスタムタグを使ってフィルタリング
Data
Overview
Data Table
Table Views
App Definitions
Data as Service
Build a Travel Planning Assistant
Travel Assistant Overview
Create a Travel Virtual Assistant
Design Conversation Skills
Create an ‘Update Booking’ Task
Create a Change Flight Task
Build a Knowledge Graph
Schedule a Smart Alert
Design Digital Skills
Configure Digital Forms
Configure Digital Views
Train the Assistant
Use Traits
Use Patterns
Manage Context Switching
Deploy the Assistant
Use Bot Functions
Use Content Variables
Use Global Variables
Use Web SDK
Build a Banking Assistant
Migrate External Bots
Google Dialogflow Bot
APIs & SDKs
API Reference
API Introduction
Rate Limits
API List
koreUtil Libraries
SDK Reference
SDK Introduction
Web SDK
How the Web SDK Works
SDK Security
SDK Registration
Web Socket Connect and RTM
Tutorials
Widget SDK Tutorial
Web SDK Tutorial
BotKit SDK
BotKit SDK Deployment Guide
Installing the BotKit SDK
Using the BotKit SDK
SDK Events
SDK Functions
Tutorials
BotKit - Blue Prism
BotKit - Flight Search Sample VA
BotKit - Agent Transfer
  1. ホーム
  2. Docs
  3. Virtual Assistants
  4. Natural Language
  5. Kore.ai XO GPT Module

Kore.ai XO GPT Module

The new Kore.ai XO GPT Models module provides fine-tuned large language models optimized for enterprise conversational AI applications. These models have been evaluated and fine-tuned to be accurate, safe, and efficient for production deployment. Initial capabilities include Conversation Summarization and User Query Rephrasing. Additional models for features like Intent Resolution, Bot Response Rephrasing, Entity Co-referencing, etc., are planned in future updates.

Note: The Kore.ai XO GPT Module is currently available only for English language interactions. The respective models are available only in our global deployment (US Cloud – https://bots.kore.ai).

Advantages of Using Kore.ai XO GPT

The following are a few advantages of using Kore.ai XO GPT.

  • Accuracy: The XO GPT module leverages smaller foundation models, typically under 10 billion parameters, that have been fine-tuned specifically for conversational AI applications. By tuning smaller models rather than directly prompting larger generative models, the XO GPT Models achieve better accuracy, relevance, and interpretability for production deployment.
  • High Performance: The XO GPT Models are hosted along with the XO Platform and are relatively smaller in size. This results in faster response times, making them suitable for production use cases for digital and voice interactions.
  • Accelerate Time-to-Value with Pre-Tuned Models: The Kore.ai XO GPT Models come pre-fine-tuned for conversational AI use cases, eliminating the complex process of prompt engineering required for adopting commercial LLMs. Enterprises can rapidly deploy these models to start realizing value immediately without needing in-house machine learning expertise or long tuning cycles.
  • Data Security and Privacy: The Kore.ai XO GPT Models are fully integrated into the XO Platform, enabling the same enterprise-grade data confidentiality, privacy, and governance enforced across the XO stack.

Fine Tuning Process

Fine-tuning is an iterative process of taking a pre-trained LLM and adapting it to perform conversational AI tasks.

Kore.ai’s Model Fine-Tuning Process

  1. Data Collection: We gather a dataset specific to the desired task. This dataset serves as the training material for fine-tuning the LLM.
  2. Selecting a base LLM: The pre-trained LLM is loaded considering their purpose and the features to which they will be applied.
  3. Training Process: The training process utilizes the task-specific dataset to train the model, involving the model’s parameters adjustment to grasp the nuances of the target task.
  4. Validation and Iteration: The fine-tuned model is evaluated on a separate validation dataset to ensure it performs well. If necessary, we iterate through the fine-tuning process to achieve optimal results.

Features Supported by Kore.ai XO GPT

The Kore.ai XO GPT module supports the following features:

  • Conversation Summary: This model generates concise, natural language summaries of interactions between the virtual assistant, users, and human agents. It distills the key intents, entities, decisions, and outcomes into an easy-to-read synopsis. Companies can leverage conversation summarization to boost agent productivity, ensure process compliance, and create better contextual recommendations – without having to read lengthy transaction histories. It is pre-integrated with Kore.ai’s Contact Center platform. It is also extensible to third-party applications via API integration. Learn more.
  • Rephrase Dialog Responses: This feature sends all User Prompts, Error Prompts, and Bot Responses to the configured LLM with the conversation context, which depends on the configured number of user inputs. Responses are rephrased based on the context and user emotion, providing the end user with a more empathetic, natural, and contextual conversation experience. Learn more.
  • Rephrase User Query: This XO GPT model utilizes the bot domain knowledge and conversation history to expand and rephrase user queries for improved understanding by downstream NLP components. This includes better recognition of contextual intents, entity co-referencing, and more. Learn more.

 

Kore.ai XO GPT Module

The new Kore.ai XO GPT Models module provides fine-tuned large language models optimized for enterprise conversational AI applications. These models have been evaluated and fine-tuned to be accurate, safe, and efficient for production deployment. Initial capabilities include Conversation Summarization and User Query Rephrasing. Additional models for features like Intent Resolution, Bot Response Rephrasing, Entity Co-referencing, etc., are planned in future updates.

Note: The Kore.ai XO GPT Module is currently available only for English language interactions. The respective models are available only in our global deployment (US Cloud – https://bots.kore.ai).

Advantages of Using Kore.ai XO GPT

The following are a few advantages of using Kore.ai XO GPT.

  • Accuracy: The XO GPT module leverages smaller foundation models, typically under 10 billion parameters, that have been fine-tuned specifically for conversational AI applications. By tuning smaller models rather than directly prompting larger generative models, the XO GPT Models achieve better accuracy, relevance, and interpretability for production deployment.
  • High Performance: The XO GPT Models are hosted along with the XO Platform and are relatively smaller in size. This results in faster response times, making them suitable for production use cases for digital and voice interactions.
  • Accelerate Time-to-Value with Pre-Tuned Models: The Kore.ai XO GPT Models come pre-fine-tuned for conversational AI use cases, eliminating the complex process of prompt engineering required for adopting commercial LLMs. Enterprises can rapidly deploy these models to start realizing value immediately without needing in-house machine learning expertise or long tuning cycles.
  • Data Security and Privacy: The Kore.ai XO GPT Models are fully integrated into the XO Platform, enabling the same enterprise-grade data confidentiality, privacy, and governance enforced across the XO stack.

Fine Tuning Process

Fine-tuning is an iterative process of taking a pre-trained LLM and adapting it to perform conversational AI tasks.

Kore.ai’s Model Fine-Tuning Process

  1. Data Collection: We gather a dataset specific to the desired task. This dataset serves as the training material for fine-tuning the LLM.
  2. Selecting a base LLM: The pre-trained LLM is loaded considering their purpose and the features to which they will be applied.
  3. Training Process: The training process utilizes the task-specific dataset to train the model, involving the model’s parameters adjustment to grasp the nuances of the target task.
  4. Validation and Iteration: The fine-tuned model is evaluated on a separate validation dataset to ensure it performs well. If necessary, we iterate through the fine-tuning process to achieve optimal results.

Features Supported by Kore.ai XO GPT

The Kore.ai XO GPT module supports the following features:

  • Conversation Summary: This model generates concise, natural language summaries of interactions between the virtual assistant, users, and human agents. It distills the key intents, entities, decisions, and outcomes into an easy-to-read synopsis. Companies can leverage conversation summarization to boost agent productivity, ensure process compliance, and create better contextual recommendations – without having to read lengthy transaction histories. It is pre-integrated with Kore.ai’s Contact Center platform. It is also extensible to third-party applications via API integration. Learn more.
  • Rephrase Dialog Responses: This feature sends all User Prompts, Error Prompts, and Bot Responses to the configured LLM with the conversation context, which depends on the configured number of user inputs. Responses are rephrased based on the context and user emotion, providing the end user with a more empathetic, natural, and contextual conversation experience. Learn more.
  • Rephrase User Query: This XO GPT model utilizes the bot domain knowledge and conversation history to expand and rephrase user queries for improved understanding by downstream NLP components. This includes better recognition of contextual intents, entity co-referencing, and more. Learn more.

 

メニュー