Chat card engine(s) for conversational chat cards

1 . A system comprising:

one or more computer readable storage media;

one or more processors operatively coupled with the one or more computer readable storage media; and

an application comprising program instructions stored on the one or more computer readable storage media that, when executed by the one or more processors, direct a computing system to at least:

receive, from a client device, an indication to start a chat card exercise;

determine, by a chat card engine, a source document based on the chat card exercise;

determine, by the chat card engine, a plurality of subtopics within the source document;

generate, by the chat card engine, a plurality of chat cards based on the source document, wherein each of the plurality of chat cards corresponds to a respective subtopic of the plurality of subtopics present in the source document;

generate, by a virtual assistant integrated with the chat card engine and configured as an artificial intelligence (AI) tutor, a first prompt based on a first subtopic;

generate, by the chat card engine via the virtual assistant, a first chat card comprising the first prompt;

receive, by the chat card engine via the virtual assistant, a first query from the client device responsive to the first prompt;

generate, by a Large Language Model (LLM) operably coupled with the chat card engine, an understanding score based on an interaction history within the first chat card, wherein the interaction history comprises at least the first query and the first prompt;

generate, by the chat card engine via the virtual assistant, a second prompt based on the understanding score; and

provide, by the chat card engine via the virtual assistant, the second prompt to the client device on the first chat card.

2 . The system of claim 1 , wherein the program instructions to generate, by the chat card engine, the plurality of chat cards based on the source document cause, when executed by the one or more processors, to further direct the computing system to:

generate, by the chat card engine, a question prompt and a corresponding answer prompt for each respective chat card within the plurality of chat cards, wherein the first prompt comprises a respective question prompt for the first chat card.

3 . The system of claim 1 , wherein the program instructions further direct the computing system to:

determine, by the chat card engine, a type of query based on the first query; and

generate, by the chat card engine, the second prompt based on the type of query and the understanding score, wherein the type of query comprises: an example query, an off-topic query, an exploratory query, and a hint request query.

4 . The system of claim 1 , wherein the program instructions further direct the computing system to:

determine, by the chat card engine, a query expansion requirement based on the first query;

retrieve, by the chat card engine, respective content from an external source based on the first query, wherein the respective content corresponds to the first subtopic and the first query;

generate, by the chat card engine, a citation to the respective content within the external source; and

generate, by the chat card engine, the second prompt based on the respective content and the understanding score, wherein the second prompt comprises the citation.

5 . The system of claim 1 , wherein the program instructions to determine, by the chat card engine, the understanding score based on the first query cause, when executed by the one or more processors, the program instructions to further direct the computing system to:

determine, by an understanding module, the understanding score based on an interaction history between the client device and the chat card engine within the first chat card and the first subtopic of the first chat card.

6 . The system of claim 1 , wherein the second prompt comprises content relating to the subtopic from the source document; and

the program instructions further direct the computing system to generate, by the chat card engine, a citation to the content within the source document.

7 . A method comprising:

receiving, from a client device, an indication to start a chat card exercise;

determining, by a chat card engine, a source document based on the chat card exercise;

determine, by the chat card engine, a plurality of subtopics within the source document;

generating, by the chat card engine, a plurality of chat cards based on the source document, wherein each of the plurality of chat cards correspond to a respective subtopic of the plurality of subtopics present in the source document;

generating, by a virtual assistant integrated with the chat card engine and configured as an artificial intelligence (AI) tutor, a first prompt based on a first subtopic;

providing, by the chat card engine via the virtual assistant, a first chat card comprising the first prompt to the client device, wherein the plurality of chat cards comprises the first chat card;

receiving, by the chat card engine via the virtual assistant, a first query from the client device responsive to the first prompt;

generating, by a Large Language Model (LLM) operably coupled with the chat card engine, an understanding score based on an interaction history within the first chat card, wherein the interaction history comprises at least the first query and the first prompt;

generating, by the chat card engine via the virtual assistant, a second prompt based on the understanding score; and

providing, by the chat card engine via the virtual assistant, the second prompt to the client device on the first chat card.

8 . The method of claim 7 , wherein:

generating, by the chat card engine, the plurality of chat cards based on the source document further comprises:

generating, by the chat card engine, a question prompt and an answer prompt for each respective chat card within the plurality of chat cards, wherein the first prompt comprises a respective question prompt for the first chat card; and

determining, by the chat card engine, the understanding score based on the first query comprises:

comparing, by the chat card engine, the first query to a corresponding answer prompt to the first prompt; and

determining, by the chat card engine, the understanding score based on the comparison of the first query to the corresponding answer prompt for the first chat card.

9 . The method of claim 7 , the method further comprising:

determining, by the chat card engine, a query expansion requirement based on the first query;

retrieving, by the chat card engine, respective content from an external source based on the first query, wherein the respective content corresponds to the first subtopic and the first query; and

generating, by the chat card engine, the second prompt based on the respective content and the understanding score.

10 . The method of claim 7 , the method further comprising:

performing, by the chat card engine, a validation process on the second prompt prior to providing the second prompt to the client device, wherein the validation process comprises a content moderation process and a schema validation process.

11 . The method of claim 7 , the method further comprising:

receiving, by the chat card engine, a hint query from the client device; and

generating, by the chat card engine, a hint responsive to the hint query, wherein the hint is based on the first subtopic and an interaction history between the client device and the chat card engine within the first chat card.

12 . The method of claim 7 , the method further comprising:

receiving, by the chat card engine, a second query from the client device responsive to the second prompt;

determining, by the chat card engine, a second understanding score based on the second query;

generating, by the chat card engine, a third prompt based on the understanding score; and

providing, by the chat card engine, the third prompt to the client device on the first chat card.

13 . The method of claim 7 , wherein the understanding score is determined to be an advancing understanding, and the method further comprises:

generating, by the chat card engine, a third prompt based on a second subtopic;

generating, by the chat card engine, a second chat card comprising the third prompt;

receiving, by the chat card engine, a second query from the client device responsive to the third prompt;

determining, by the chat card engine, a second understanding score based on the second query;

generating, by the chat card engine, a fourth prompt based on the second understanding score; and

providing, by the chat card engine, the fourth prompt to the client device on the second chat card.

14 . The method of claim 7 , the method further comprising:

determining, by the chat card engine, an understanding score for each of the plurality of chat card; and

generating, by the chat card engine, a grade for the chat card exercise based on the understanding score for each of the plurality of chat cards within the chat card exercise.

15 . A computer readable storage media comprising processor-executable instructions configured to cause one or more processors to:

receive, from a client device, an indication to start a chat card exercise;

determine, by a chat card engine, a source document based on the chat card exercise;

determine, by the chat card engine, a plurality of subtopics within the source document;

generate, by the chat card engine, a plurality of chat cards based on the source document, wherein each of the plurality of chat cards corresponds to a respective subtopic of the plurality of subtopics present in the source document;

generate, by a virtual assistant integrated with the chat card engine and configured as an artificial intelligence (AI) tutor, a first prompt based on a first subtopic;

generate, by the chat card engine via the virtual assistant, a first chat card comprising the first prompt;

receive, by the chat card engine via the virtual assistant, a first query from the client device responsive to the first prompt;

generate, by a Large Language Model (LLM) operably coupled with the chat card engine, an understanding score based on an interaction history within the first chat card, wherein the interaction history comprises at least the first query and the first prompt;

generate, by the chat card engine via the virtual assistant, a second prompt based on the understanding score; and

provide, by the chat card engine via the virtual assistant, the second prompt to the client device on the first chat card;

generate, by the chat card engine via the virtual assistant, a second prompt based on the understanding score; and

provide, by the chat card engine via the virtual assistant, the second prompt to the client device on the first chat card.

16 . The computer readable storage media of claim 15 , wherein:

the processor-executable instructions to generate, by the chat card engine, the plurality of chat cards based on the source document cause the one or more processors to further execute processor-executable instructions stored in the computer readable storage media to:

generate, by the chat card engine, a question prompt and an answer prompt for each respective chat card within the plurality of chat cards, wherein the first prompt comprises a respective question prompt for the first chat card; and

the processor-executable instructions to determine, by the chat card engine, the understanding score based on the first query cause the one or more processors to further execute processor-executable instructions stored in the computer readable storage media to:

compare, by the chat card engine, the first query to a corresponding answer prompt to the first prompt;

determine, by the chat card engine, a query expansion requirement based on the comparison of the first query to the corresponding answer prompt for the first chat card;

retrieve, by the chat card engine, respective content from an external source based on the first query based on the query expansion requirement, wherein the respective content corresponds to the first subtopic and the first query; and

compare, by the chat card engine, the first query to the respective content from the external source.

17 . The computer readable storage media of claim 15 , wherein the processor-executable instructions cause the one or more processors to further execute processor-executable instructions stored in the computer readable storage media to:

perform, by the chat card engine, a validation process on the second prompt prior to providing the second prompt to the client device, wherein the validation process comprises one or more quality metrics.

18 . The computer readable storage media of claim 15 , wherein the processor-executable instructions cause the one or more processors to further execute processor-executable instructions stored in the computer readable storage media to:

determine, by the chat card engine, an understanding score for each of the plurality of chat cards within the chat card exercise; and

determine, by the chat card engine, a grade the chat card exercise based on the understanding score for each of the plurality of chat cards.

19 . The computer readable storage media of claim 15 , wherein:

the processor-executable instructions cause the one or more processors to further execute processor-executable instructions stored in the computer readable storage media to:

determine, by the chat card engine, that the first query is an off-topic query; and

the processor-executable instructions to generate, by the chat card engine, the second prompt cause the one or more processors to further execute processor-executable instructions stored in the computer readable storage media to:

generate, by the chat card engine, a redirecting prompt, wherein the redirecting prompt comprises content relating to the first subtopic.

20 . The computer readable storage media of claim 15 , wherein the processor-executable instructions cause the one or more processors to further execute processor-executable instructions stored in the computer readable storage media to:

determine, by the chat card engine, a query expansion requirement based on the first query;

retrieve, by the chat card engine, respective content from an external source based on the first query, wherein the respective content corresponds to the first subtopic and the first query; and

generate, by the chat card engine, the second prompt based on the respective content and the understanding score.