DingTalk AI Assistant Upgrade: Multi-Modality, Long Text, and RPA Empowering Work and Productivity

On March 28, DingTalk AI Assistant announced a series of significant upgrades, including product capabilities such as image comprehension, document speed reading, and workflow, marking a major breakthrough in the integration of multi-modal, long text, and RPA (Robotic Process Automation) technologies into AI applications. Enhanced Visual Reasoning and Long Text Reading Capabilities Powered by Alibaba's Tongyi Thousand Questions Large Model Based on the Tongyi Thousand Questions large model developed by Alibaba, the upgraded DingTalk AI Assistant boasts stronger visual reasoning and long text reading capabilities

DingTalk AI Assistant Upgrade: Multi-Modality, Long Text, and RPA Empowering Work and Productivity

On March 28, DingTalk AI Assistant announced a series of significant upgrades, including product capabilities such as image comprehension, document speed reading, and workflow, marking a major breakthrough in the integration of multi-modal, long text, and RPA (Robotic Process Automation) technologies into AI applications.

Enhanced Visual Reasoning and Long Text Reading Capabilities Powered by Alibaba's Tongyi Thousand Questions Large Model

Based on the Tongyi Thousand Questions large model developed by Alibaba, the upgraded DingTalk AI Assistant boasts stronger visual reasoning and long text reading capabilities.

Image Comprehension: An All-in-One "Image Encyclopedia"

Empowered by the Tongyi Thousand Questions Qwen-VL-Max visual comprehension model, DingTalk AI Assistant can accurately describe and recognize image information, and perform operations such as information inference, extended creation, text extraction, and translation based on the image, making it an all-in-one "image encyclopedia."

Document Speed Reading: Rapidly Grasp File Content

The Tongyi Thousand Questions large model grants DingTalk AI Assistant exceptional document speed reading capabilities. Users can simply send local files (such as Word, PDF, PPT, Excel, etc.), DingTalk documents, or web links of various formats, and the AI Assistant will swiftly parse and extract keywords, generating an intelligent summary. Even for video content up to 2GB in size, the AI Assistant can complete intelligent interpretation within 3 minutes.

Workflow Capability Supported by RPA Technology

With the Agent (intelligent entity) technology popularized by OpenAI, the integration of large models into automation technologies has become a widely accepted direction. DingTalk AI Assistant has introduced workflow capabilities in this upgrade, enabling the AI Assistant to perform a series of more complex tasks.

Workflow: An Advanced Way to Utilize AI Assistant

Workflow is an advanced application of AI Agent, allowing users to decompose and arrange the process of AI task execution during creation, so that the AI Assistant can proactively take over and complete the corresponding operations. In addition, the workflow can connect to external system data and API capabilities, further expanding the AI Assistant's range of actions, such as building a creative AI Assistant capable of automatically writing scripts and generating videos.

Lowering the Threshold for Use: Convenient Workflow Templates

To reduce the threshold for user adoption, DingTalk provides a variety of workflow templates. Enterprise users can utilize workflow templates to create a store information collection assistant, automatically organize user feedback, and store it in a DingTalk multi-dimensional table, helping employees save on trivial tasks; individual users can also create assistants that automatically track hot topics and write articles by connecting to the Weibo API, executing commands in batches, and significantly improving content production efficiency.

Seamless Integration of Multi-Modality, Long Text, and Workflow Functions

Currently, users can seamlessly utilize the multi-modal, long text, and workflow functions of AI Assistant within the DingTalk APP or PC client. Whether it's sending long files, online documents, web links, or video content in the IM chat box, or accessing the AI Assistant dialog box via the magic wand button, users can effortlessly experience the powerful capabilities of the AI Assistant, performing operations such as person recognition, location recognition, analysis, answering, translation, summarization, text extraction, and even engaging in intelligent Q&A through multi-round interactions.

The latest upgrade of DingTalk AI Assistant marks another major breakthrough in DingTalk's application of AI technologies. The integration of multi-modality, long text, and RPA technologies enables the AI Assistant to comprehensively understand and process different types of information, execute more complex tasks, and assist enterprises and individuals in improving their work and productivity. As DingTalk AI Assistant continues to refine and evolve, we anticipate it unleashing greater potential in the future, delivering a more intelligent and efficient work experience for users.


Disclaimer: The content of this article is sourced from the internet. The copyright of the text, images, and other materials belongs to the original author. The platform reprints the materials for the purpose of conveying more information. The content of the article is for reference and learning only, and should not be used for commercial purposes. If it infringes on your legitimate rights and interests, please contact us promptly and we will handle it as soon as possible! We respect copyright and are committed to protecting it. Thank you for sharing.(Email:[email protected])