Author Aide API is Learnosity’s AI-powered Item generator and translator, aiming to supercharge your Authors' productivity.

Author Aide API is available as a standalone API to give customers greater flexibility for integration. Author Aide API can be accessed independently of any other Learnosity API. To use Author Aide API, you will need a valid subscription added to your Item bank, which gives you credits to use in Author Aide. Credits are used to save generated or translated items. Author Aide can also be used with other Learnosity-powered authoring tools like Author Site or Author API.  

With Author Aide, authors can create Items containing different types of Learnosity Questions such as multiple choice, true/false or cloze drag-and-drop. Authors only need to add basic prompts like topic, subject, grade, etc. to generate Items. However, note that the more context you provide, the higher the expected quality of your output will be.

Once the Items are generated or translated, authors can review and save the ones they are happy with to their item bank. If a single item is generated or translated, authors can edit the Item before saving it into their Item bank. Authors can refine the generated Questions further by adding their own instructions before saving or even change their earlier prompts to start fresh.

 

Overview of uploading a file, specifying the criteria and reviewing the generated questions

 

 

Generate single or bulk item(s) with Author Aide

In the following section, you will find a breakdown of the input options provided by Author Aide. Depending on your subscription, you may or may not have access to all of these features. 

While these inputs provide one layer of model interaction, we continually improve our prompting strategies. We have implemented additional safeguards to ensure that generated items are unbiased, respectful, and equitable. Author Aide will prevent content generation and explain if prompts are mismatched or might produce inaccurate or biased results.

Hint: The more prompts Authors add during their Item generation, the more refined and better the generated Item will be.

 

User interface showcasing the different sections discussed in further detail below

 

A) Define the context of your items

In the first section, you will find the basic information the model needs to understand the context of the intended items. 

How many questions do you want?

By specifying the number of items, you can bulk generate up to 25 items at once. The quality of the output is heavily reliant on the quality of the input. Providing limited source material or files and requesting numerous questions may yield lower-quality results. For optimal results, ensure that your input is comprehensive and relevant, especially when requesting numerous items.

What is the point value for each item? 

Provide the maximum score for each item generated. 

Paste (source material)

This field can be used to paste in a specific passage or source material that should be used to generate the items. This content will remain private and not be used to train the underlying AI model. Images and multimedia are not yet accepted, but tabular data is accepted (CSV or text-based).

File upload (RAG)

With this feature, authors can upload any PDF file with a maximum size of 100MB to a secure and private database. Files are linked to an author and cannot be shared across multiple authors. Any text-based information will be extracted from the file, while images or multimedia without text will be ignored. Each file upload will require a unique title, which will be saved and searchable via the search field. All text-based content from the uploaded files will be used for generation purposes when selected. Hence, the provided inputs must match the content within the file, else Author Aide will not be able to extract relevant content from the file to generate Items. For example, if a topic is specified, it should exist in the file's content. 

Do you have a specific scenario in mind?

This input further focuses the model with three options:

Option 1: Specific topic (I have a topic in mind)

Setting a topic will provide the model with a specific focus. If source material (or file) has been provided, the topic must match this context. The topic can be general, very specific or even include specific skills, depending on your needs. 

Some examples:

• “Velocity in space” is a general statement that may provide a variety of different questions
• “Role of earthworms in soil conservation” is more specific and will likely create a set of more focused questions 
• “Test student's vocabulary focusing on sports as a topic” also defines the skillset that the student needs to demonstrate 
• “Problem-solving and data analysis” may be the focus you want to set when you have used a CSV in the source material 

Option 2: Exact question (I have an exact question in mind)

This input only supports a single generation. The contents entered here will be retained and used to inform the response options and distractors that are generated. The more specific you are, the better the quality of the output will be. 

Some examples:

• “A spacecraft, named Explorer A, is travelling through space on a research mission. If the spacecraft has covered a distance of 300,000 kilometres in 5 hours, what is its velocity?” 
• “Given the following data set representing measurements of the Iris flower, identify which of the following statements is true:

| ID | Sepal Length | Sepal Width | Petal Length | Petal Width |

|----|--------------|-------------|--------------|-------------|

| 1 | 5.1 | 3.5 | 1.4 | 0.2 | 

| 2 | 4.9 | 3.0 | 1.4 | 0.2 | 

| 3 | 4.7 | 3.2 | 1.3 | 0.2 |

| 4 | 4.6 | 3.1 | 1.5 | 0.2 | 

| 5 | 5.0 | 3.6 | 1.4 | 0.2 |

Which of the following statements is correct based on the data”

Option 3: Custom prompt (I would like to use my own prompt)

This input allows the author to create a prompt that best suits their needs. For example, if you need to measure a specific skill set, have a style guide to follow or even have a style of question in mind. Note that all other inputs will also be added, and you can still include a topic (or focus), you will just have to spell it out. 

Some examples:

• Create different question styles: “Create a scenario-based question involving three kids named Jack, Jill and Pete travelling in a spacecraft. The question should test the learner's knowledge of gravity in space”
• Generate data sets or passages alongside your question: “Create a question that focuses on testing a student’s data interpretation and analysis skills using the topic of plants. As part of the question, provide a HTML table that is based on the topic. The response options should provide different interpretations of the table.” 
• Define a specific style needed: “Use the source material to generate a question that tests the student's ability to reason. Ensure that the response options are numbered, for example, "1." and that they are just one sentence without any full stops. Include the source material in the stimulus field and as part of the question.” 

None

It is possible to generate items without specifying a specific topic, question or prompt when the source material has been pasted above. In this case, questions will be based on the source material provided rather than accessing publicly available information.

For file uploads, you must specify a focus (topic, question, or custom prompt) to generate items. This is necessary because files alone are usually too large and lack the required direction for effective generation.

B) Specify the criteria

To further improve the quality of generated items and ensure they meet institutional criteria, you can provide additional context in the second section. 

Add learning objective or standard

It is highly recommended to fill in this input, as it provides the model with a specific goal for your items. This is especially crucial when targeting a particular student's skillset or expecting a specific skill demonstration.

Some examples:

• A publicly available standard: “ “MS-PS1-4 standard of the Next Generation Science Standards”
• A full description of a standard: “MS-PS1-4. Develop a model that predicts and describes changes in particle motion, temperature, and phase (state) of a substance when thermal energy is added or removed.”
• Create your own learning objective: “Critical Thinking and Analysis: Students should exercise critical thinking skills by analyzing experimental data, recognizing patterns, and making logical conclusions about scientific inquiries.”

Subject

The subject field provides the focus for a particular area of knowledge and offers flexibility, extending beyond K12 to include subjects in higher education and vocational programs.

Some examples:

• K12 subject examples: Biology, Physical Education, World History
• Higher education examples: Quantum Physics, Philosophy, Culinary Arts
• Vocational examples: Automotive Technology, Medical Billing, Welding Technology

Grade

The grade field applies to K12 education only and can be left blank if it doesn't fit your assessment needs. This input helps Author Aide understand the age and developmental stage of the learners, which impacts the reading level and complexity of the generated items. If you need a reading level that differs from the specified grade, use the custom prompt option.

Level of Bloom’s taxonomy & Depth of knowledge

Item creation commonly utilizes Bloom's Taxonomy and Depth of Knowledge (DOK). Bloom's Taxonomy addresses the type of cognitive processing (such as analysis or synthesis), whereas DOK focuses on the task's depth and complexity (e.g., Level 4, extended thinking). These frameworks can be used in conjunction. If you employ different educational frameworks, define the expected learning level or criteria using the custom prompt option in the first section. Not all framework levels may be optimal for all question types, use these at your discretion.

C) Setup your responses

Depending on the type of question being generated, the inputs in the last section will be different. This section usually determines the number of response options and whether feedback, such as distractor rationale, is included in the generation.

Note that Author Aide will make its best attempt at generating the requested number of Items but its underlying model capabilities are restricted in terms of complexities involved with generating a large number of questions with a large number of answer options. For example, if the maximum possible responses are selected for a topic, the expected number of questions may not be achievable. 

Translate item(s) with Author Aide

Effortlessly create multilingual content with Author Aide's translation feature. This powerful tool allows you to translate up to 20 items simultaneously, with each item containing up to 50 widgets. Simply select your desired target language and review the completed translations. For your convenience and to facilitate efficient searches within your item bank, the titles of the items remain in the original language.

Specify a target language and choose a set of items for translation

 

Review generated items

Single generation and translation 

After the generation of a single question has been completed, the author will be presented with an opportunity to refine the question or adjust their original inputs. The refinement will take the given question and adjust it per the instructions. 

Some examples:

• Add an introductory paragraph at the beginning of the question 
• Change the question to a scenario-style question 
• Number each response option with “1.”, “2.”, etc. 
• Add in the correct answer to the distractor's rational 

Bulk generation and translation

After the items have been generated, they will be displayed within the "Review generated items" list and remain available for 30 days. The review process empowers authors to engage with the generated content by evaluating the rationale behind the distractors and examining both correct/incorrect response options. Translated questions from your items are also compared against the original questions, as shown below.

After review, authors can add the items to their item bank. Upon addition to the item bank, these items transition into an editable state, granting authors the flexibility to modify and customize them according to their specific requirements.

Provide feedback 

We welcome your feedback to improve the quality of generated items and our overall user experience. We invite you to share your thoughts by completing this survey, which will be sent directly to our team.

Please note that we do not track any personal information through this survey. If you want to report a bug or official request, please contact your customer service representative or the support team.