サインイン
はじめてのオートメーションをわずか数分で作成できます。Studio Web を試す →
Prompt Engineering Framework
作成者: UiPath
0
ソリューション
172
![[戻る] ボタン](https://marketplace.uipath.com/static/images/icons/react-slick-active/arrow_back_slick.svg){kind=icon}
![[戻る] ボタン](https://marketplace.uipath.com/static/images/icons/react-slick-active/arrow_back_slick_mobile.svg){kind=icon}
![[次へ] ボタン](https://marketplace.uipath.com/static/images/icons/react-slick-active/arrow_next_slick.svg){kind=icon}
![[次へ] ボタン](https://marketplace.uipath.com/static/images/icons/react-slick-active/arrow_next_slick_mobile.svg){kind=icon}
要約
要約
Enhance your prompts using collected context, and continuously adjust Generative AI parameters and prompts to tailor LLM outcomes to retrieve the desired output.
概要
概要
Does your organization deal with vast text-based operations? Perhaps it's crafting cold-call emails, summarizing customer interactions, processing insurance claims communications, creating product documentation, or responding to customer feedback. These are just a few examples. Any operation that involves heavy use of text, across various systems, document types, and applications, can be daunting and time-consuming.
The Prompt Engineering Framework Solution seamlessly takes in any number of textual contexts, enabling a Prompt Engineer to refine the prompt and retrieve accurate responses from a Large Language Model (LLM). The iteration process involves the modification of prompts, AI parameters, and even model changes, until the desired outcome is achieved. Once the Prompt Engineer is satisfied with the quality, accuracy, and reliability of the LLM's output, they can choose to run the refined prompt across all items in a queue without needing to verify each response manually. This streamlines the process significantly, saving your team the time spent on individual manual verification. With this confidence in the process, they can focus more on refining the prompt and adjusting Generative AI parameters to consistently produce high-quality responses.
Two practical applications of this framework are available as Accelerators on the Marketplace named:
機能
機能
- Human-in-the-loop validation of LLM responses. Users can change prompts, parameters, models, and providers from a single verification screen.
- Automatically insert context into a prompt.
- Automatically run against all context in a queue or manually verify each LLM response.
その他の情報
その他の情報
依存関係
• UiPath.Form.Activities: [2.0.3] • UiPath.GoogleVertex.IntegrationService.Activities: [5.0.0-preview] • UiPath.MicrosoftAzureOpenAI.IntegrationService.Activities: [6.1.0] • UiPath.OpenAI.IntegrationService.Activities: [6.1.0] • UiPath.Persistence.Activities: [1.4.3] • UiPath.SolutionAccelerators.Reusables.Generic: [1.1.0]
コード言語
Visual Basic
技術情報
バージョン
1.0.0更新日
2023年7月18日対応製品
Studio: 23.4+
認定
ゴールド 認定済み
コレクション
サポート
メール アドレス: solution-accelerators-support@uipath.com
リンク: https://marketplace.uipath.com/contact-usSupport on Demand is available only to UiPath Enterprise customers. To access this type of support, you need to have an active UiPath license. More details are available here: https://docs.uipath.com/marketplace/automation-cloud/latest/user-guide/support-on-demand. Note: Marketplace Support On Demand is not subject to a response SLA, but one should expect a response within two business days.
リソース
類似のコンポーネント
