1 code implementation • EMNLP 2021 • Junyan Cheng, Iordanis Fostiropoulos, Barry Boehm, Mohammad Soleymani
We evaluate our model with three sentiment analysis datasets and achieve comparable or superior performance compared with the existing methods, with a 90% reduction in the number of parameters.
no code implementations • 3 Aug 2023 • Junyan Cheng, Peter Chin
Bridging the huge disparity between neural and symbolic representation can potentially enable the incorporation of symbolic thinking into neural networks from essence.
1 code implementation • 1 Dec 2021 • Junyan Cheng, Iordanis Fostiropoulos, Barry Boehm
The fusion between a graph representation like Abstract Syntax Tree (AST) and a source code sequence makes the use of current approaches computationally intractable for large input sequence lengths.
1 code implementation • 17 Nov 2021 • Junyan Cheng, Iordanis Fostiropoulos, Barry Boehm
SCG is the result of the early fusion between a source code snippet and the AST representation.
no code implementations • 1 Jan 2021 • Junyan Cheng, Iordanis Fostiropoulos, Barry Boehm
As opposed to natural languages, source code understanding is influenced by grammar relations between tokens regardless of their identifier name.