Generating and utilizing models for long-range event relation extraction

A system comprising:

• one or more memory devices comprising an event relation extraction model and a short-range event relation dataset;
• and
• one or more processors configured to cause the system to:
• access the short-range event relation dataset comprising a digital document comprising an event pair that includes a first event within a first host sentence of the digital document and a second event within a second host sentence within the digital document and the first host sentence and the second host sentence are within a short-range threshold of each other;
• determine a current number of sentences separating the first host sentence and the second host sentence;
• determine a minimum constraint comprising a number of synthetic sentences to insert based on a difference between a long-range event relation threshold and the current number of sentences;
• transform the short-range event relation dataset to induce long-range event relations within the short-range event relation dataset by generating, utilizing a generative language model, a set of synthetic sentences based on the determined number of synthetic sentences to insert based on the difference between the long-range event relation threshold and the current number of sentences;
• insert the set of synthetic sentences within the digital document across a plurality of insertion locations between the first host sentence and the second host sentence that are initially separated by the short-range threshold to separate the event pair by the number of synthetic sentences determined based on the difference between the long-range event relation threshold and the current number of sentences;
• generate a long-range event relation dataset by augmenting the digital document within the short-range event relation dataset to include the set of synthetic sentences comprising the number of synthetic sentences between the first host sentence and the second host sentence;
• modify the event relation extraction model to determine long-range event relations from digital documents by learning model parameters for the event relation extraction model from the long-range event relation dataset;
• and
• generate, utilizing the event relation extraction model, an event relation graph indicating a relationship between a long-range event pair, wherein the event relation graph comprises nodes representing events and edges representing relationships between nodes.

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

• accessing a digital document comprising a first sentence and a second sentence separated by a number of sentences that satisfies a long-range event relation threshold;
• determining a long-range event pair that includes the first sentence and the second sentence from the digital document, by utilizing an event relation extraction model comprising parameters learned from:
• accessing a training short-range event relation dataset;
• determining a current number of sentences separating a first host sentence and a second host sentence in the training short-range event relation dataset;
• determining a minimum constraint comprising a number of synthetic sentences to insert based on a difference between a long-range event relation threshold and the current number of sentences;
• and
• transforming the training short-range event relation dataset to induce long-range event relations by generating, utilizing a generative language model, a set of synthetic sentences based on the determined number of synthetic sentences to insert based on the difference between the long-range event relation threshold and the current number of sentences;
• and
• generating an event relation graph indicating a long-range event relation between the long-range event pair, wherein the event relation graph comprises nodes representing events and edges representing relationships between nodes.