Automated item listing generation system

1 . A system comprising:

a computer-readable memory comprising a machine learning model; and

one or more processors in communication with the memory, wherein computer-executable instructions, when executed by the one or more processors, cause the one or more processors to at least:

generate information to provide an interactive graphical user interface;

receive a request, via the interactive graphical user interface, to generate an item listing for an item, wherein the request further comprises an indication of a content platform;

retrieve a content seed comprising item information associated with the item;

generate the item listing based in part on providing the content seed as input to the machine learning model, wherein the item listing is formatted according to an object format associated with the content platform, and wherein the item listing is generated in a single pass, wherein to generate the item listing in a single pass, an input is provided to the machine learning model resulting in an output through a single unit of execution of the machine learning model without the machine learning model retaining a previous input; and

store the item listing in a database associated with a commerce platform, wherein storing the item listing in the database causes an item associated with the item listing to be presented by the commerce platform as available for commerce.

2 . The system of claim 1 , wherein the machine learning model is configured to generate item listings according to the object format.

3 . The system of claim 1 , wherein the computer-executable instructions, when executed, further cause the one or more processors to:

retrieve an example item listing formatted according to the object format;

provide the example item listing as further input to the machine learning model, wherein the item listing is generated based in part on the example item listing.

4 . The system of claim 1 , wherein the machine learning model is a large language model.

5 . A computer-implemented method comprising:

under control of a computing device comprising one or more processors configured to execute specific instructions,

receiving a request to generate an item listing;

retrieving a content seed;

providing the content seed as input to a machine learning model;

obtaining the item listing based in part on an output of the machine learning model generated in response to the content seed being provided as input to the machine learning model, wherein the item listing is generated in an object format, and wherein the item listing is generated in a single pass of the machine learning model such that the content seed being provided as input to the machine learning model results in generation of the output through a single execution unit of the machine learning model; and

transmitting the item listing.

6 . The computer-implemented method of claim 5 , further comprising:

identifying an object store storing a plurality of item listings associated with a content platform indicated in the request, wherein the content seed is retrieved from the object store.

7 . The computer-implemented method of claim 5 , wherein the single execution unit of the machine learning model occurs without the machine learning model retaining a previous input.

8 . The computer-implemented method of claim 5 , further comprising:

identifying an object store storing a plurality of existing item listings for an online platform indicated by the request, wherein the plurality of existing item listings is in the object format; and

retrieving an existing item listing of the plurality of existing item listings from the object store, wherein the machine learning model further generates the item listing in response to being provided the existing item listing.

9 . The computer-implemented method of claim 5 , further comprising:

identifying an object store storing a plurality of existing item listings for an online platform indicated by the request, wherein the plurality of existing item listings are in the object format;

identifying a content seed store storing a plurality of existing content seeds associated with the plurality of existing item listings;

providing at least one content seed of the plurality of existing content seeds as input to the machine learning model;

obtaining at least one training output based in part on the output of the machine learning model generated in response to the at least one content seed being provided as input to the machine learning model;

comparing the at least one training output to at least one existing item listing of the plurality of existing item listings to generate a comparison result; and

updating the machine learning model based on the comparison result, wherein the result of updating the machine learning model is to train the machine learning model to generate item listings in the object format.

10 . The computer-implemented method of claim 5 , further comprising:

identifying a machine learning model store where the machine learning model indicated by the request is stored, wherein the machine learning model is configured to generate item listings in the object format;

retrieving the machine learning model from the machine learning model store.

11 . The computer-implemented method of claim 5 , further comprising:

determining the content seed is missing an information item used by the machine learning model to generate the item listing in the object format, wherein generating the at least one training output by the machine learning model comprises inferring a value for the information item based at least in part on latent knowledge of the machine learning model.

12 . The computer-implemented method of claim 5 , further comprising:

processing the content seed to reduce a total amount of information of the content seed, wherein the total amount of information of the content seed is reduced to a minimum amount of information needed by the machine learning model to generate the item listing.

13 . A non-transitory machine-readable storage medium encoded with instructions executable by a processor of a computing device, wherein the instructions, when executed by the processor, cause the computing device to at least:

receive a request to generate an item listing;

retrieve a content seed;

provide the content seed as input to a machine learning model;

obtain the item listing based in part on an output of a machine learning model generated in response to the content seed being provided as input to the machine learning model, wherein the item listing is generated in an object format, and wherein the item listing is generated in a single pass of the machine learning model such that the content seed being provided as input to the machine learning model results in generation of the output through a single execution unit of the machine learning model; and

transmit the item listing.

14 . The non-transitory machine-readable storage medium of claim 13 , wherein the machine learning model is a large language model.

15 . The non-transitory machine-readable storage medium of claim 13 , wherein the request comprises a uniform resource locator (URL), and wherein the content seed comprises item information retrieved based on the URL.

16 . The non-transitory machine-readable storage medium of claim 13 , wherein the content seed is at least one of: a PDF, an image, a video, a database record, and/or text.

17 . The non-transitory machine-readable storage medium of claim 13 , wherein the content seed is retrieved by a first retriever, and wherein the instructions, when executed by the processor, further cause the computing device to:

retrieve a second content seed using a second retriever; and

wherein the second content seed is provided as input to the machine learning model at a same time as the content seed.

18 . The non-transitory machine-readable storage medium of claim 13 , wherein the instructions, when executed by the processor, further cause the computing device to:

search, using a second machine learning model, for item information associated with the requested item listing;

based on the search, retrieve item information from a plurality of item information sources; and

wherein the content seed comprises the retrieved item information.

19 . The non-transitory machine-readable storage medium of claim 13 , wherein the instructions, when executed by the processor, further cause the computing device to:

search, using a second machine learning model, for item information associated with the content seed;

based on the search, retrieve item information from a plurality of item information sources; and

modify the content seed to include the retrieved item information.

20 . The non-transitory machine-readable storage medium of claim 13 , wherein the instructions, when executed by the processor, further cause the computing device to:

process, by a second machine learning model, the content seed, wherein to process the content seed the second machine learning model converts the content seed to an information format accepted as input by the machine learning model.