Generating presentation slides with distilled content

1. A method for generating presentation slides with distilled content, comprising:

providing a user interface (UI) for presentation to a user, the UI displaying a slide generation UI element allowing the user to request slide generation using content of at least a portion of a data file as source material;

receiving, via the UI, a user selection of the slide generation UI element;

identifying a plurality of logical breakpoints in the content based on at least one of a format or size of a plurality of content items of the content;

determining, based on the identified plurality of logical breakpoints, a set of slides to be included in a slide presentation;

identifying a layout template for each of the set of slides based on the plurality of content items of the content;

applying a trained machine learning model to the plurality of content items of the content to obtain an output of the trained machine learning model, the output of the trained machine learning model indicating one or more distilled content items and a format for generating a presentation visualization item comprising the one or more distilled content items, wherein the one or more distilled content items include a subset of the plurality of content items of the content, and wherein the trained machine learning model is trained using training data comprising a set of training files and corresponding summaries for the training files to learn what text segment to include in a distilled content item and what format to use for a presentation visualization item comprising the distilled content item; and

generating the set of slides based on each identified layout template and the presentation visualization item, the presentation visualization item having the indicated format.

2. The method of claim 1 , wherein the UI displays the data file having a plurality of portions and allows the user to select a first portion of the plurality of portions of the data file, wherein the plurality of content items of the content is associated with the first portion of the data file.

3. The method of claim 2 , further comprising:

receiving an indication of a user selection of a second portion of the data file; and

distilling a second plurality of content items of the second portion of the data file into one or more additional distilled content items to generate a second presentation visualization item, wherein each of the one or more additional distilled content items includes a subset of the second plurality of content items of the second portion of the data file, and wherein at least one slide of the set of slides is further generated based on the second presentation visualization item.

4. The method of claim 1 , wherein the training data comprises (i) training inputs comprising the set of training files and (ii) corresponding target outputs comprising the corresponding summaries for the training files, wherein a corresponding target output further comprises a format for a summary of a respective training file.

5. The method of claim 1 , wherein the plurality of content items of the content comprises a first set of sentences, and the one or more distilled content items comprise a second set of sentences that includes fewer sentences than the first set of sentences, and wherein the generated presentation visualization item comprises a list that is based on the second set of sentences.

6. The method of claim 1 , wherein the plurality of content items of the content comprises a data table, and the one or more distilled content items comprise a range of data from the data table, and wherein the generated presentation visualization item comprises a data chart that is based on the range of data.

7. The method of claim 1 , wherein the plurality of content items of the content comprises an image, and wherein the presentation visualization item comprises the image.

8. The method of claim 1 , further comprising:

receiving an interaction with the set of slides; and

using the interaction for a heuristic rule to be applied to generation of a subsequent set of slides.

9. The method of claim 1 , further comprising:

setting text of a parent header in the data file as a title for the respective layout template of the slide; and

setting the one or more distilled content items including text associated with the parent header as a body for the layout template.

10. A system to generate presentation slides with distilled content, the system comprising:

a memory; and

a processing device, coupled to the memory, to perform operations comprising:

providing a user interface (UI) for presentation to a user, the UI displaying a slide generation UI element allowing the user to request slide generation using content of at least a portion of a data file as source material;

receiving, via the UI, a user selection of the slide generation U I element;

identifying a plurality of logical breakpoints in the content based on at least one of a format or size of a plurality of content items of the content;

determining, based on the identified plurality of logical breakpoints, a set of slides to be included in a slide presentation;

identifying a layout template for each of the set of slides based on the plurality of content items of the content;

applying a trained machine learning model to the plurality of content items of the content to obtain an output of the trained machine learning model, the output of the trained machine learning model indicating one or more distilled content items and a format for generating a presentation visualization item comprising the one or more distilled content items, wherein the one or more distilled content items include a subset of the plurality of content items of the content, and wherein the trained machine learning model is trained using training data comprising a set of training files and corresponding summaries for the training files to learn what text segment to include in a distilled content item and what format to use for a presentation visualization item comprising the distilled content item; and

generating the set of slides based on each identified layout template and the presentation visualization item, the presentation visualization item having the indicated format.

11. The system of claim 10 , wherein the training data comprises (i) training inputs comprising the set of training files and (ii) corresponding target outputs comprising the corresponding summaries for the training files, wherein a corresponding target output further comprises a format for a summary of a respective training file.

12. The system of claim 10 , wherein the plurality of content items of the content comprises a first set of sentences, and the one or more distilled content items comprise a second set of sentences that includes fewer sentences than the first set of sentences, and wherein the generated presentation visualization item comprises a list that is based on the second set of sentences.

13. The system of claim 10 , wherein the plurality of content items of the content comprises a data table, and the one or more distilled content items comprise a range of data from the data table, and wherein the generated presentation visualization item comprises a data chart that is based on the range of data.

14. The system of claim 10 , wherein the plurality of content items of the content comprises an image, and wherein the presentation visualization item comprises the image.

15. The system of claim 10 , the operations further comprising:

receiving an interaction with the set of slides; and

using the interaction for a heuristic rule to be applied to generation of a subsequent set of slides.

16. The system of claim 10 , the operations further comprising:

setting text of a parent header in the data file as a title for the respective layout template of the slide; and

setting the one or more distilled content items including text associated with the parent header as a body for the layout template.

17. A non-transitory, computer-readable medium storing instructions that, when executed by a processing device, cause the processing device to perform operations comprising:

providing a user interface (UI) for presentation to a user, the UI displaying a slide generation UI element allowing the user to request slide generation using content of at least a portion of a data file as source material;

receiving, via the UI, a user selection of the slide generation UI element;

identifying a plurality of logical breakpoints in the content based on at least one of a format or size of a plurality of content items of the content;

determining, based on the identified plurality of logical breakpoints, a set of slides to be included in a slide presentation;

identifying a layout template for each of the set of slides based on the plurality of content items of the content;

applying a trained machine learning model to the plurality of content items of the content to obtain an output of the trained machine learning model, the output of the trained machine learning model indicating one or more distilled content items and a format for generating a presentation visualization item comprising the one or more distilled content items, wherein the one or more distilled content items include a subset of the plurality of content items of the content, and wherein the trained machine learning model is trained using training data comprising a set of training files and corresponding summaries for the training files to learn what text segment to include in a distilled content item and what format to use for a presentation visualization item comprising the distilled content item; and

generating the set of slides based on each identified layout template and the presentation visualization item, the presentation visualization item having the indicated format.

18. The non-transitory, computer-readable medium of claim 17 , wherein the training data comprises (i) training inputs comprising the set of training files and (ii) corresponding target outputs comprising the corresponding summaries for the training files, wherein a corresponding target output further comprises a format for a summary of a respective training file.

19. The non-transitory, computer-readable medium of claim 17 , wherein the plurality of content items of the content comprises a first set of sentences, and the one or more distilled content items comprise a second set of sentences that includes fewer sentences than the first set of sentences, and wherein the generated presentation visualization item comprises a list that is based on the second set of sentences.

20. The non-transitory, computer-readable medium of claim 17 , wherein the plurality of content items of the content comprises a data table, and the one or more distilled content items comprise a range of data from the data table, and wherein the generated presentation visualization item comprises a data chart that is based on the range of data.