Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

pointer-generator network

A pointer-generator network is a neural sequence-to-sequence architecture that combines the ability to generate new words from a fixed vocabulary with the ability to copy words directly from the input text. At each decoding step, the model computes a generation probability that acts as a soft switch, balancing between predicting tokens from its predefined vocabulary distribution and pointing to specific tokens in the source sequence via an attention distribution. This hybrid mechanism allows the network to accurately reproduce out-of-vocabulary words, rare terms, and specific factual details from the source while preserving grammatical fluency. To mitigate repetitive outputs, the architecture is frequently augmented with a coverage mechanism that tracks previously attended source tokens.

1 item

RACE: Retrieval-augmented Commit Message Generation

RACE: Retrieval-augmented Commit Message Generation

Ensheng Shi, Yanlin Wang, Wei Tao, Lun Du, Hongyu Zhang, Shi Han, Dongmei Zhang, Hongbin Sun

Why you should read this

Proposes a retrieval-augmented generation framework that uses an exemplar guider to control the influence of retrieved historical commits, significantly improving the quality and accuracy of automated commit messages across multiple programming languages.

Commit messages are important for developers to understand changes in code repositories. However, writing commit messages is a time-consuming and tedious task for developers. To alleviate this burden, many approaches have been proposed to automatically generate commit messages. Among them, the state-of-the-art approaches are based on neural machine translation (NMT) models. However, these approaches only utilize the difference between the pre- and post-commit versions of the code (i.e., code diff) to generate commit messages, and ignore the rich information in the commit history. In this paper, we propose a novel approach named RACE (Retrieval-augmented Commit message gEneration) that retrieves similar code diffs and their corresponding commit messages from the commit history, and leverages the retrieved commit messages to guide the generation of commit messages. We evaluate RACE on a large-scale dataset collected from GitHub. Experimental results show that RACE outperforms the state-of-the-art approaches by a significant margin.

Added

2026-10-02