Granted Patent B2

US 12,566,916 · App. 18/169,802 · Granted Mar 3, 2026

Inventors: Maria Iu (Berkeley, CA); Lakshman Somasundaram (San Francisco, CA); Yilin Li (Jersey City, NJ); Shweta Patira (New York, NY); Adam Kaplan (Brooklyn, NY); Sara Remi Fields (Brooklyn, NY)

Assignee: Microsoft Technology Licensing, LLC

G06F40/197G06F40/117G06F40/166G06F40/258G06N3/0475G06N3/08G06Q10/101G06F8/33H04L67/06

Abstract

Claims (87)

1 . A method comprising:

generating a display of a first version of a first document, wherein the first version of the first document comprises a plurality of segments of digital content generated by a generative language model;

receiving a first contribution comprising digital content associated with a first user of a network;

creating a first segment-contribution pair by linking the first contribution with a first segment of the plurality of segments;

positioning the first segment-contribution pair at a first GUI position within the display of the first version of the first document;

receiving a second contribution comprising digital content output by the generative language model based on input to the generative language model from the first user or a second user of the network;

creating a second segment-contribution pair by linking the second contribution with a second segment of the plurality of segments;

positioning the second segment-contribution pair at a second GUI position within the display of the first version of the first document;

based on social action data relating to at least one of the first contribution or the second contribution, re-positioning the first segment-contribution pair from the first GUI position to the second GUI position or re-positioning the second segment-contribution pair from the second GUI position to the first GUI position;

generating a second version of the first document comprising at least one of the first contribution or the second contribution; and

publishing the second version of the first document to the network.

2 . The method of claim 1 , further comprising:

generating a prompt for the generative language model based on the at least one of the first segment-contribution pair or the second segment-contribution pair.

3 . The method of claim 1 , further comprising:

linking a third user input GUI control element with the first contribution; and

receiving, via the network and the third user input GUI control element, first social action data relating to the first contribution.

4 . The method of claim 1 , further comprising:

linking a first GUI position with the first segment;

receiving, via the network, social action data relating to the first contribution; and

based on the social action data, assigning the first contribution to the first GUI position.

5 . The method of claim 1 , further comprising:

receiving, via the network, first social action data relating to the first contribution; and

based on the first social action data, outputting, by the generative language model, a first version of a second document, wherein the first version of the second document is based on at least one of the first segment or the first contribution.

6 . The method of claim 1 , further comprising:

receiving, via the network, first social action data relating to the first contribution;

based on the first social action data, generating training data, wherein the training data is based on the first segment-contribution pair; and

training the generative language model based on the training data.

7 . The method of claim 1 , further comprising:

receiving, via the network, first social action data relating to the first contribution; and

based on the first social action data, linking a badge with the first user.

8 . The method of claim 1 , further comprising:

identifying the plurality of segments by searching the first version of the first document for header text.

9 . The method of claim 8 , further comprising:

identifying header text in the first version of the first document;

identifying, in the first version of the first document, digital content related to the header text; and

creating a segment comprising the header text and the identified digital content related to the header text.

10 . The method of claim 1 , further comprising:

retrieving, from at least one data store, skill set data associated with the first user; and

linking the retrieved skill set data with the first contribution.

11 . A system comprising:

at least one processor; and

at least one memory coupled to the at least one processor;

wherein the at least one memory comprises instructions which, when executed by the at least one processor, cause the at least one processor to perform operations comprising:

generating a display of a first version of a first document, wherein the first version of the first document comprises a plurality of segments of digital content generated by a generative language model;

receiving a second contribution comprising digital content output by the generative language model based on input to the generative language model from a user of a network;

creating a first segment-contribution pair by linking the first contribution with a first segment of the plurality of segments;

positioning the first segment-contribution pair at a first GUI position within the display of the first version of the first document;

receiving a second contribution comprising digital content associated with the first user or a second user of the network;

creating a second segment-contribution pair by linking the second contribution with a second segment of the plurality of segments;

positioning the second segment-contribution pair at a second GUI position within the display of the first version of the first document;

based on social action data relating to at least one of the first contribution or the second contribution, re-positioning the first segment-contribution pair from the first GUI position to the second GUI position or re-positioning the second segment-contribution pair from the second GUI position to the first GUI position;

generating a second version of the first document comprising at least one of the first contribution or the second contribution; and

publishing the second version of the first document to the network.

12 . The system of claim 11 , wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations further comprising:

generating a prompt for the generative language model based on the at least one of the first segment-contribution pair or the second segment-contribution pair.

13 . The system of claim 11 , wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations further comprising:

linking a first GUI position with the first segment;

receiving, via the network, first social action data relating to the first contribution; and

based on the first social action data, assigning the first contribution to the first GUI position.

14 . The system of claim 11 , wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations further comprising:

receiving, via the network, first social action data relating to the first contribution; and

based on the first social action data, outputting, by the generative language model, a first version of a second document, wherein the first version of the second document is based on at least one of the first segment or the first contribution.

15 . The system of claim 11 , wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations further comprising:

receiving, via the network, first social action data relating to the first contribution;

based on the first social action data, generating training data, wherein the training data is based on the first segment-contribution pair; and

training the generative language model based on the training data.

16 . The system of claim 11 , wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations further comprising:

receiving, via the network, the social action data relating to at least one of the first contribution or the second contribution; and

assigning the plurality of segments to GUI positions based on the social action data.

17 . The system of claim 11 , wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations further comprising:

identifying header text in the first version of the first document;

identifying, in the first version of the first document, digital content related to the header text; and

creating a segment comprising the header text and the identified digital content related to the header text.

18 . The system of claim 11 , wherein the instructions, when executed by the at least one processor, cause the at least one processor to perform operations further comprising:

retrieving, from at least one data store, skill set data associated with the first user; and

linking the retrieved skill set data with the first contribution.

19 . A non-transitory computer-readable medium comprising instructions that when executed by a processor, cause the processor to:

generate a display of a first version of a first document, wherein the first version of the first document comprises a plurality of segments of digital content generated by a generative language model;

receive a first contribution comprising digital content associated with a first user of a network;

create a first segment-contribution pair by linking the first contribution with a first segment of the plurality of segments;

position the first segment-contribution pair at a first GUI position within the display of the first version of the first document;

receive a second contribution comprising digital content output by the generative language model based on input to the generative language model from the first user or a second user of the network;

create a second segment-contribution pair by linking the second contribution with a second segment of the plurality of segments;

position the second segment-contribution pair at a second GUI position within the display of the first version of the first document;

based on social action data relating to at least one of the first contribution or the second contribution, re-position the first segment-contribution pair from the first GUI position to the second GUI position or re-positioning the second segment-contribution pair from the second GUI position to the first GUI position;

generate a second version of the first document comprising at least one of the first contribution or the second contribution; and

publish the second version of the first document to the network.

Assignments (1)

Continuity (1)

References Cited (146)