SYSTEMS AND METHODS FOR TRAINING LANGUAGE MODEL PARAMETERS TO ACCESS DATA SOURCES IN A DISTRIBUTED COMPUTING ENVIRONMENT
Systems and methods for training a language model using historical data structures are disclosed. A system can maintain, in one or more data structures, data corresponding to a plurality of data structures. The system can generate, using the data corresponding to the plurality of historical data structures, a training dataset comprising a plurality of training examples. At least one training example can include a respective input prompt indicating an intent relating to data structures and a respective output message identifying information associated with at least one historical data structure of the plurality of historical data structures. The system can update a language model using the training dataset.
1 . A system, comprising:
one or more processors coupled to non-transitory memory, the one or more processors configured to:
maintain, in one or more data structures, data corresponding to a plurality of historical wager opportunities;
generate, using the data corresponding to the plurality of historical wagers, a training dataset comprising a plurality of training examples, at least one training example comprising:
(i) a respective input prompt indicating an intent relating to wager opportunities, and
(ii) a respective output message identifying information associated with at least one historical wager of the plurality of historical wager opportunities; and
update a language model using the training dataset.
2 . The system of claim 1 , wherein the one or more processors are further configured to:
generate a training example of the plurality of training examples wherein the respective input prompt of the training example indicates a request for information relating to a live event, and the respective output message of the training example indicates the information relating to the live event to satisfy the request.
3 . The system of claim 2 , wherein the respective input prompt further comprises data corresponding to the live event retrieved from one or more databases.
4 . The system of claim 1 , wherein the one or more processors are further configured to:
generate a training example of the plurality of training examples wherein the respective input prompt of the training example indicates a request for odds information, and the respective output message of the training example indicates odds of at least one historical wager of the plurality of historical wagers.
5 . The system of claim 1 , wherein the one or more processors are further configured to:
generate a training example of the plurality of training examples wherein the respective input prompt of the training example indicates a request for location-based wager information, and the respective output message of the training example indicates at least one historical wager corresponding to a location indicated in the input prompt.
6 . The system of claim 1 , wherein the one or more processors are further configured to:
generate a training example of the plurality of training examples wherein the respective input prompt of the training example indicates a request for profile information, and the respective output message of the training example indicates at least a subset of data stored in association with a player profile.
7 . The system of claim 1 , wherein the one or more processors are further configured to:
generate a training example of the plurality of training examples that includes a plurality of input prompts and a plurality of output messages in a conversation format.
8 . The system of claim 1 , wherein the one or more processors are further configured to:
generate a training example of the plurality of training examples wherein the respective output message of the training example indicates a classification of intent of the respective input prompt.
9 . The system of claim 8 , wherein the classification of the intent corresponds to one or more of a request for a wager recommendation, a request to modify a wager, a request for information relating to a live event, a request for information relating to at least one wager opportunity, a request to place a wager, or a request for information maintained by the one or more processors.
10 . The system of claim 1 , wherein the one or more processors are further configured to:
update the language model using one or more of a supervised training process or a self-supervised training process.
11 . A method, comprising:
maintaining, by one or more processors coupled to non-transitory memory, in one or more data structures, data corresponding to a plurality of historical wager opportunities;
generating, by the one or more processors, using the data corresponding to the plurality of historical wagers, a training dataset comprising a plurality of training examples, at least one training example comprising:
(i) a respective input prompt indicating an intent relating to wager opportunities, and
(ii) a respective output message identifying information associated with at least one historical wager of the plurality of historical wager opportunities; and
updating, by the one or more processors, a language model using the training dataset.
12 . The method of claim 11 , further comprising:
generating, by the one or more processors, a training example of the plurality of training examples wherein the respective input prompt of the training example indicates a request for information relating to a live event, and the respective output message of the training example indicates the information relating to the live event to satisfy the request.
13 . The method of claim 12 , wherein the respective input prompt further comprises, by the one or more processors, data corresponding to the live event retrieved from one or more databases.
14 . The method of claim 11 , further comprising:
generating, by the one or more processors, a training example of the plurality of training examples wherein the respective input prompt of the training example indicates a request for odds information, and the respective output message of the training example indicates odds of at least one historical wager of the plurality of historical wagers.
15 . The method of claim 11 , further comprising:
generating, by the one or more processors, a training example of the plurality of training examples wherein the respective input prompt of the training example indicates a request for location-based wager information, and the respective output message of the training example indicates at least one historical wager corresponding to a location indicated in the input prompt.
16 . The method of claim 11 , further comprising:
generating, by the one or more processors, a training example of the plurality of training examples wherein the respective input prompt of the training example indicates a request for profile information, and the respective output message of the training example indicates at least a subset of data stored in association with a player profile.
17 . The method of claim 11 , further comprising:
generating, by the one or more processors, a training example of the plurality of training examples that includes a plurality of input prompts and a plurality of output messages in a conversation format.
18 . The method of claim 11 , further comprising:
generating, by the one or more processors, a training example of the plurality of training examples wherein the respective output message of the training example indicates a classification of intent of the respective input prompt.
19 . The method of claim 18 , wherein the classification of the intent corresponds to one or more of a request for a wager recommendation, a request to modify a wager, a request for information relating to a live event, a request for information relating to at least one wager opportunity, a request to place a wager, or a request for information maintained by the one or more processors.
20 . The method of claim 11 , further comprising:
updating, by the one or more processors, the language model using one or more of a supervised training process or a self-supervised training process.