Employing abstract meaning representation to lay the last mile towards reading comprehension

1. A method of verifying an alignment of text, the method comprising:

identifying a plurality of common subtrees between a first semantic tree that represents a first text and a second semantic tree that represents a second text, wherein each semantic tree comprises semantic nodes that represent entities and semantic edges that represent relationships between two of the entities, wherein a common subtree comprises (i) nodes, each node representing a common entity that is common between the first semantic tree and the second semantic tree and (ii) edges between the nodes that represent a semantic relationship between the entities;

calculating a semantic alignment score from a sum of sizes of each of the plurality of common subtrees, wherein a size of a common subtree equals a number of nodes in the common subtree;

identifying, between a first syntactic tree that represents the first text and a second syntactic tree that represents the second text, a number of common syntactic nodes, wherein each syntactic tree comprises syntactic nodes that represent a word and an associated part of speech;

calculating a syntactic alignment score based on the number of common syntactic nodes; and

responsive to determining that a sum of the semantic alignment score and the syntactic alignment score is greater than a threshold, outputting the second text to a device.

2. The method of claim 1 , wherein the identifying the number of common syntactic nodes comprises:

applying a trained classification model to the first syntactic tree and the second syntactic tree; and

receiving, from the trained classification model, information indicating common syntactic nodes.

3. The method of claim 1 , further comprising:

determining an additional semantic alignment score between the first text and an additional text;

determining an additional syntactic alignment score between the first text and the additional text; and

responsive to determining that an additional sum of the additional semantic alignment score and the additional syntactic alignment score is greater than a threshold, outputting the additional text to the device.

4. The method of claim 1 , further comprising:

providing the first text into a classification model, wherein the classification model is trained to select a body of text corresponding to the first text; and

obtaining the second text from the classification model.

5. The method of claim 4 , further comprising training the classification model by iteratively:

providing, to the classification model, training text and a corresponding reference answer;

receiving, from the classification model, a generated answer;

comparing entities in the generated answer to entities in the reference answer; and

adjusting, based on the comparing, a parameter of the classification model to minimize a loss function.

6. The method of claim 1 , further comprising obtaining one or more of the first text or the second text from a body of text.

7. The method of claim 1 , further comprising:

translating additional text into a set of vectors, wherein each vector represents words and sentences from the additional text;

providing the set of vectors to a classification model, wherein the classification model is trained to generate answers from the additional text; and

obtaining, from the classification model, the second text.

8. A system comprising:

a non-transitory computer-readable medium storing computer-executable program instructions; and

a processing device communicatively coupled to the non-transitory computer-readable medium for executing the computer-executable program instructions, wherein executing the computer-executable program instructions configures the processing device to perform operations comprising:

identifying a plurality of common subtrees between a first semantic tree that represents a first text and a second semantic tree that represents a second text, wherein each semantic tree comprises semantic nodes that represent entities and semantic edges that represent relationships between two of the entities, wherein a common subtree comprises (i) nodes, each node representing a common entity that is common between the first semantic tree and the second semantic tree and (ii) edges between the nodes that represent a semantic relationship between the entities;

calculating a semantic alignment score from a sum of sizes of each of the plurality of common subtrees, wherein a size of a common subtree equals a number of nodes in the common subtree;

identifying, between a first syntactic tree that represents the first text and a second syntactic tree that represents the second text, a number of common syntactic nodes, wherein each syntactic tree comprises syntactic nodes that represent a word and an associated part of speech;

calculating a syntactic alignment score based on the number of common syntactic nodes; and

responsive to determining that a sum of the semantic alignment score and the syntactic alignment score is greater than a threshold, outputting the second text to a device.

9. The system of claim 8 , wherein the identifying comprises:

applying a trained classification model to the first syntactic tree and the second syntactic tree; and

obtaining information indicating common syntactic nodes from the classification model.

10. The system of claim 8 , wherein executing the computer-executable program instructions configures the processing device to perform operations comprising:

determining an additional semantic alignment score between the first text and an additional text;

determining an additional syntactic alignment score between the first text and the additional text; and

responsive to determining that an additional sum of the additional semantic alignment score and the additional syntactic alignment score is greater than a threshold, outputting the additional text to the device.

11. The system of claim 8 , wherein executing the computer-executable program instructions configures the processing device to perform operations comprising:

providing the first text into a classification model, wherein the classification model is trained to select a body of text corresponding to the first text; and

obtaining, from the classification model, the second text.

12. The system of claim 11 , wherein executing the computer-executable program instructions configures the processing device to perform operations comprising training the classification model by iteratively:

providing, to the classification model, a training question and text comprising context corresponding to the training question;

receiving, from the classification model, an answer;

comparing entities in answer to entities in a reference answer that corresponds to the training question; and

adjusting, based on the comparing, a parameter of the classification model to minimize a loss function.

13. The system of claim 8 , wherein executing the computer-executable program instructions configures the processing device to perform operations comprising:

translating additional text into a set of vectors, wherein each vector represents words and sentences from the additional text;

providing the set of vectors to a classification model, wherein the classification model is trained to generate answers from the additional text; and

obtaining, from the classification model, the second text.

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

accessing a first semantic tree for a first text and second semantic tree for a second text, wherein each semantic tree comprises nodes and edges, wherein the nodes represent entities, and wherein each edge represents a relationship between two of the entities;

identifying, between the first semantic tree and the second semantic tree, a plurality of common subtrees, wherein a common subtree comprises (i) nodes, each node representing a common entity that is common between the first semantic tree and the second semantic tree and (ii) edges between the nodes that represent a semantic relationship between the entities;

calculating a semantic alignment score from a sum of sizes of each of the plurality of common subtrees, wherein a size of a common subtree equals a number of nodes in the common subtree; and

responsive to determining that the semantic alignment score is greater than a threshold, outputting the second text to a device.

15. The non-transitory computer-readable storage medium of claim 14 , wherein when executed by a processing device, the program instructions cause the processing device to perform operations comprising:

providing the first text into a classification model, wherein the classification model is trained to select a body of text corresponding to the first text; and

obtaining, from the classification model, the second text.

16. The non-transitory computer-readable storage medium of claim 15 , wherein when executed by a processing device, the program instructions cause the processing device to perform operations comprising training the classification model by iteratively:

providing, to the classification model, a training question and text comprising context corresponding to the training question;

receiving, from the classification model, an answer;

comparing entities in answer to entities in a reference answer that corresponds to the training question; and

adjusting, based on the comparing, a parameter of the classification model to minimize a loss function.

17. The non-transitory computer-readable storage medium of claim 14 , wherein when executed by a processing device, the program instructions cause the processing device to perform operations comprising:

determining a syntactic alignment score between the first text and an additional text; and

responsive to determining that a sum of the semantic alignment score and the syntactic alignment score is greater than a threshold, outputting the additional text to the device.

18. The non-transitory computer-readable storage medium of claim 14 , wherein when executed by a processing device, the program instructions cause the processing device to perform operations comprising:

translating additional text into a set of vectors, wherein each vector represents words and sentences from the additional text;

providing the set of vectors to a classification model, wherein the classification model is trained to generate answers from the additional text; and

obtaining, from the classification model, the second text.

19. The non-transitory computer-readable storage medium of claim 18 , wherein the translating comprises providing the first text and the second text to a word2vec model.

20. The non-transitory computer-readable storage medium of claim 14 , wherein the first text represents a question and the second text represents an answer.