WRSWorld Research StateContribute
WORLD RESEARCH STATE

GraphCodeBERT: data-flow-aware masked code pretraining

Contribution grade

Structured code representation learning · Human-designed research method or resource

Reference score120,000Provisional editorial estimate of this exact contribution; its grade is assessed separately.

What changed

Added extracted variable data-flow nodes to source/comment inputs, used graph-guided attention masks and trained code-edge and variable/token alignment tasks alongside masked language modeling.

Scientific scope

Added extracted variable data-flow nodes to source/comment inputs, used graph-guided attention masks and trained code-edge and variable/token alignment tasks alongside masked language modeling. Graph extraction depends on parsers and supported constructs; learned attention is not complete semantic analysis or proof of program equivalence. Data-flow analysis and Transformers are inherited.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Graph extraction depends on parsers and supported constructs; learned attention is not complete semantic analysis or proof of program equivalence. Data-flow analysis and Transformers are inherited.

Contributors

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction