
Creating high-quality educational materials often requires thorough evaluation and flexible content organization. To meet these needs, we’ve introduced an updated evaluation interface in EdFlow, offering a clearer view of item quality and a more refined approach to how evaluations are conducted. We’ve also enhanced the passage interactive mode, making it easier to work with longer texts.
Presets now support images, giving you more options for presenting information visually. Meanwhile, we’re laying the groundwork for datasets integration in the agent builder, ensuring you have access to the most up-to-date and relevant information for your educational content.
A key improvement is the smart tagging system in the content generation flow. Instead of manually assigning tags, EdFlow automatically determines the most appropriate ones from your organization’s existing tags, saving you time and helping maintain consistency across your items.
By refining evaluation methods, broadening content formats, and automating data and tag management, EdFlow aims to help you maintain a consistent standard of educational content while reducing administrative overhead.

Attaching media to your educational content should be straightforward, but sometimes the process can feel cumbersome. To make this easier, we've enhanced the media selection and upload flow in EdWrite.
Here's what's new:
Select or Upload Media: You can now choose existing media from your library or upload new files directly within the CMS. This simplifies adding images, videos, and other media to your items.
Searchable Media Library: The media selection feature now includes a search function, allowing you to quickly find the media you need by searching through your library.
These updates make managing and attaching media files more convenient, helping you focus on creating engaging educational content.

Understanding how your tags and content are interconnected can be challenging when dealing with complex educational materials. To help with this, we've released the Knowledge Graph UI version 1.0 in EdWrite.
Graphical Representation of Tags Hierarchy: The Knowledge Graph UI provides a visual map of your tags and how they relate to each other. This helps you see the structure of your content at a glance.
This new feature gives you better insight into your content organization, making it easier to manage and navigate your educational materials.

Finding specific items among a vast collection of educational content can be like searching for a needle in a haystack. To make this task easier, we've enhanced the data grid UI in EdWrite.
The improvements include:
Improved Filters UI helps you locate specific items more easily, allowing you to refine your searches with greater precision.
Views Support enables you to save your preferred filters and sorting options for quick access. For example, you can create a view that displays all items currently "In Review."
Show/Hide Columns in your data grid, so you see only the information that's important to you.
Pin Key Columns as you scroll through large datasets, ensuring critical information is always at hand.
Additionally, we've added a Media Data Grid that allows you to view all media items available in your library. This feature provides a centralized place to browse and manage images, videos, and other media assets directly within EdWrite.
These updates are designed to help you organize and manage your data and media more effectively during content creation.

Manually entering metadata can be a tedious distraction from your main work. To address this, we've introduced dynamic metadata attributes in EdWrite. Now, when you add new educational content, the system automatically generates metadata fields for form inputs based on your content. This means you can focus on creating materials without handling repetitive data entry.
We've also updated the data grid to support these dynamic metadata attributes, allowing you to view and manage metadata directly within the grid. This makes organizing and editing your content more convenient.
In addition, we've given EdWrite a fresh look with a redesigned user interface. The updated layout organizes tools and options logically, making navigation more straightforward so you can find what you need without hassle.
These improvements simplify your content management process, letting you concentrate on developing valuable educational materials.

Creating diverse educational content is important for engaging different learning styles. To support this, we've expanded EdWrite to include passage-type content alongside multiple-choice questions. For example, when developing reading comprehension sections, you can now write passages directly in EdWrite and link related questions to them.
We've also integrated AI agents—Generator and Evaluator— into the CMS to assist with content creation and assessment.
Generator: Helps you generate new items, such as passages or questions, based on the topics you provide. This can speed up the initial drafting process.
Evaluator: Allows you to evaluate existing items and receive suggestions for improving the content. It reviews your material for clarity, relevance, and alignment with learning objectives.
These additions aim to help you produce a wider range of high-quality content more efficiently, enhancing both the creation and refinement of your educational materials.

Keeping track of revisions and collaborating on educational content can be challenging. To help with this, we've introduced item versioning in EdWrite. You can now view past versions of items and roll back to previous changes if needed, giving you the flexibility to experiment while maintaining a safety net.
We've also added changelogs to track what has changed in the current version of an item. This makes it easier to see updates and understand the evolution of your content.
Additionally, we've introduced support for comments. You can add individual comments on items or when moving them to the "CHANGES_REQUESTED" status. This allows you to specify exactly what needs to be modified, making collaboration clearer and more effective.
These features give you greater control over your content and enhance the collaborative process, making it more transparent and manageable.

High-quality training data is essential for developing effective AI agents. EdFlow now provides Datasets, allowing you to curate collections of data specifically for agent training. By tagging examples as "good" or "bad," you help the agent learn from both positive and negative instances.
For instance, when reviewing agent outputs, you might find responses that meet your criteria. Tagging these as "good" includes them in the training data as positive examples. If an output doesn't meet your standards, tagging it as "bad" adds it as a negative example, teaching the agent to avoid similar mistakes in the future.
By building datasets with both positive and negative examples, you can improve your agents' performance based on actual interactions and your specific guidelines.