Generating and using joint representations of source code

1. A method implemented using one or more processors, comprising:

applying tokens of a source code snippet as input across a sequence-processing machine learning model to generate a plurality of token embeddings;

generating a graph based on the source code snippet;

augmenting the graph by infusing nodes of the graph that correspond to the tokens of the source code snippet with corresponding token embeddings of the generated plurality of token embeddings;

processing the augmented graph using a graph neural network to generate a joint representation of the source code snippet, the joint representation comprising an aggregate node embedding based on the augmented graph infused with the corresponding token embeddings, wherein the aggregate node embedding is obtained by further infusing the nodes of the augmented graph with data indicative of the corresponding token embeddings from neighboring nodes for a number of iterations by propagating the date from each node along one or more edges to one or more immediate neighboring nodes;

calculating one or more distances in latent space between the joint representation generated from the source code snippet and one or more other joint representations generated from one or more other source code snippets to make a determination about the source code snippet; and

automatically generating a summary, review, or comment associated with the determination about the source code snippet.

2. The method of claim 1 , wherein the joint representation comprises a mean over the aggregate node embeddings generated from the augmented graph.

3. The method of claim 1 , wherein the calculating is performed as part of a search for other source code similar to the source code snippet.

4. The method of claim 1 , wherein the determination about the source code snippet comprises a source code quality score.

5. The method of claim 1 , wherein the graph comprises an abstract syntax tree.

6. The method of claim 1 , wherein the sequence-processing machine learning model comprises a neural network that includes a self attention mechanism.

7. The method of claim 6 , wherein the neural network comprises a transformer network.

8. A system comprising one or more processors and memory storing instructions that, in response to execution of the instructions by the one or more processors, cause the one or more processors to:

apply tokens of a source code snippet as input across a sequence-processing machine learning model to generate a plurality of token embeddings;

generate a graph based on the source code snippet;

augment the graph by infusing nodes of the graph that correspond to the tokens of the source code snippet with corresponding token embeddings of the generated plurality of token embeddings;

process the augmented graph using a graph neural network to generate a joint representation of the source code snippet, the joint representation comprising an aggregate node embedding based on the augmented graph infused with the corresponding token embeddings, wherein the aggregate node embedding is obtained by further infusing the nodes of the augmented graph with data indicative of the corresponding token embeddings from neighboring nodes for a number of iterations by propagating the data from each node along one or more edges to one or more immediate neighboring nodes;

calculate one or more distances in latent space between the joint representation generated from the source code snippet and one or more other joint representations generated from one or more other source code snippets to make a determination about the source code snippet; and

automatically generate a summary, review, or comment associated with the determination about the source code snippet.

9. The system of claim 8 , wherein the determination comprises a similarity between the source code snippet and other source code.

10. The system of claim 8 , wherein the determination about the source code snippet comprises a source code quality score.

11. The system of claim 8 , wherein the sequence-processing machine learning model comprises a transformer network.

12. At least one non-transitory computer-readable medium comprising instructions that, in response to execution of the instructions by one or more processors, cause the one or more processors to:

apply tokens of a source code snippet as input across a sequence-processing machine learning model to generate a plurality of token embeddings;

generate a graph based on the source code snippet;

augment the graph by infusing nodes of the graph that correspond to the tokens of the source code snippet with corresponding token embeddings of the generated plurality of token embeddings;

process the augmented graph using a graph neural network to generate a joint representation of the source code snippet, the joint representation comprising an aggregate node embedding based on the augmented graph infused with the corresponding token embeddings, wherein the aggregate node embedding is obtained by further infusing the nodes of the augmented graph with data indicative of the corresponding token embeddings from neighboring nodes for a number of iterations by propagating the data from each node along one or more edges to one or more immediate neighboring nodes;

calculate one or more distances in latent space between the joint representation generated from the source code snippet and one or more other joint representations generated from one or more other source code snippets to make a determination about the source code snippet; and

automatically generate a summary, review, or comment associated with the determination about the source code snippet.