# Learning to Update Natural Language Comments Based on Code Changes

arXiv: https://arxiv.org/abs/2004.12169 · HTML: https://mlsanity.com/paper/2004.12169/

## Key facts

- Cited by 31 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Apr 2020.

## Built on by (most cited first, 30 of 31)

- [From Word Models to World Models: Translating from Natural Language to the Probabilistic Language of Thought](https://mlsanity.com/paper/2306.12672/) (2023)
- [On the Evaluation of Neural Code Summarization](https://mlsanity.com/paper/2107.07112/) (2021)
- [CoditT5: Pretraining for Source Code and Natural Language Editing](https://mlsanity.com/paper/2208.05446/) (2022)
- [NL-EDIT: Correcting semantic parse errors through natural language interaction](https://mlsanity.com/paper/2103.14540/) (2021)
- [Learning Deep Semantics for Test Completion](https://mlsanity.com/paper/2302.10166/) (2023)
- [RACE: Retrieval-Augmented Commit Message Generation](https://mlsanity.com/paper/2203.02700/) (2022)
- [Deep Learning for Code Intelligence: Survey, Benchmark and Toolkit](https://mlsanity.com/paper/2401.00288/) (2023)
- [Learning Structural Edits via Incremental Tree Transformations](https://mlsanity.com/paper/2101.12087/) (2021)
- [Source Code Summarization with Structural Relative Position Guided Transformer](https://mlsanity.com/paper/2202.06521/) (2022)
- [TypeT5: Seq2seq Type Inference using Static Analysis](https://mlsanity.com/paper/2303.09564/) (2023)
- [Impact of Evaluation Methodologies on Code Summarization](https://mlsanity.com/paper/2108.09619/) (2021)
- [Coeditor: Leveraging Contextual Changes for Multi-round Code Auto-editing](https://mlsanity.com/paper/2305.18584/) (2023)
- [Multilingual Code Co-Evolution Using Large Language Models](https://mlsanity.com/paper/2307.14991/) (2023)
- [Natural Language Outlines for Code: Literate Programming in the LLM Era](https://mlsanity.com/paper/2408.04820/) (2024)
- [Code Comment Inconsistency Detection with BERT and Longformer](https://mlsanity.com/paper/2207.14444/) (2022)
- [ContextCov: Deriving and Enforcing Executable Constraints from Agent Instruction Files](https://mlsanity.com/paper/2603.00822/) (2026)
- [Learning to Generate Code Comments from Class Hierarchies](https://mlsanity.com/paper/2103.13426/) (2021)
- [Learning to Describe Solutions for Bug Reports Based on Developer Discussions](https://mlsanity.com/paper/2110.04353/) (2021)
- [exLong: Generating Exceptional Behavior Tests with Large Language Models](https://mlsanity.com/paper/2405.14619/) (2024)
- [ComSum: Commit Messages Summarization and Meaning Preservation](https://mlsanity.com/paper/2108.10763/) (2021)
- [GN-Transformer: Fusing Sequence and Graph Representation for Improved Code Summarization](https://mlsanity.com/paper/2111.08874/) (2021)
- [A Survey on Natural Language Processing for Programming](https://mlsanity.com/paper/2212.05773/) (2022)
- [CCRep: Learning Code Change Representations via Pre-Trained Code Model and Query Back](https://mlsanity.com/paper/2302.03924/) (2023)
- [Why Does CLAUDE.md Keep Growing? Catastrophic Remembering in Agentic Coding](https://mlsanity.com/paper/2608.11095/) (2026)
- [TokDrift: When LLM Speaks in Subwords but Code Speaks in Grammar](https://mlsanity.com/paper/2510.14972/) (2025)
- [SkillScriptBench: Benchmarking Self-Evolution of Executable Agent Skill Packages Beyond Markdown](https://mlsanity.com/paper/2610.04008/) (2026)
- [Data-Driven AI Model Signal-Awareness Enhancement and Introspection](https://mlsanity.com/paper/2111.05827/) (2021)
- [LAMNER: Code Comment Generation Using Character Language Model and Named Entity Recognition](https://mlsanity.com/paper/2204.09654/) (2022)
- [RefactorAssist: Agentic Refinement for Reliable Code Refactoring](https://mlsanity.com/paper/2608.00924/) (2026)
- [Excision Score: Evaluating Edits with Surgical Precision](https://mlsanity.com/paper/2510.21537/) (2025)

## Abstract (arXiv)

We formulate the novel task of automatically updating an existing natural language comment based on changes in the body of code it accompanies. We propose an approach that learns to correlate changes across two distinct language representations, to generate a sequence of edits that are applied to the existing comment to reflect the source code modifications. We train and evaluate our model using a dataset that we collected from commit histories of open-source software projects, with each example consisting of a concurrent update to a method and its corresponding comment. We compare our approach against multiple baselines using both automatic metrics and human evaluation. Results reflect the challenge of this task and that our model outperforms baselines with respect to making edits.