Rethinking Training Data for Generating Code Review Comments

• URL: http://arxiv.org/abs/2607.25851v1
• Date: Tue, 28 Jul 2026 15:24:30 GMT
• Title: Rethinking Training Data for Generating Code Review Comments
• Abstract summary: We argue that problematic training instances are not homogeneous and that some limitations stem from deeper issues in the task formulation itself.<n>We derive a taxonomy of misalignment capturing semantic ambiguity, lack of actionability, and context dependence.
• Score: 1.7212147528714599
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Generating code review comments has become a prominent research direction in automated code review, commonly formulated as a text generation task over diff-comment pairs. Despite advances in learning-based approaches, generated review comments are often generic, weakly grounded, or non-actionable. Recent studies have also shown that review comment datasets contain noisy or unsuitable training instances, motivating LLM-based dataset cleaning approaches. In this paper, we argue that problematic training instances are not homogeneous and that some limitations stem from deeper issues in the task formulation itself. Through an empirical inspection of a widely used review comment dataset, we identify misaligned training pairs: instances where the relationship between the code change and the review comment does not provide a reliable learning signal for generating actionable review feedback from localized inputs. We derive a taxonomy of misalignment capturing three recurring sources: semantic ambiguity, lack of actionability, and context dependence. We further explore whether incorporating this taxonomy into LLM-based filtering improves the identification of problematic training instances, observing that detecting misaligned training pairs remains challenging. Based on these observations, we argue that improving review comment generation requires more than dataset cleaning alone, motivating explicit validity criteria, richer contextual inputs, and evaluation practices aligned with review intent and actionability.

Related papers

• ActReview: Rebuttal-Guided Training Data and Rubric Rewards for Actionable Peer Review Generation [54.82704239643649]
  We introduce ActReview, a rebuttal-guided post-training framework that connects paper-specific diagnoses to concrete, grounded revision plans.<n>Our central insight is that author rebuttals reveal plausible actions for addressing reviewer concerns and can therefore provide latent supervision for revision-oriented feedback.<n>We also introduce ActReview-Bench, a human-curated benchmark of 1,000 instances for evaluating diagnostic quality and revision usefulness.
  arXiv  Detail & Related papers  (2026-09-08T17:30:16Z)
• On the Impact of Code Comments for Automated Bug-Fixing: An Empirical Study [3.8571904945643336]
  Comments may play a critical role in fixing certain types of bugs by providing valuable design and implementation insights.<n>We investigate how the presence or absence of comments, both during training and at inference time, impacts the bug-fixing capabilities of Large Language Models.<n>Our findings show that comments improve ABF accuracy by up to threefold when present in both phases, while training with comments does not degrade performance when instances lack them.
  arXiv  Detail & Related papers  (2026-01-30T15:07:27Z)
• CoCoNUTS: Concentrating on Content while Neglecting Uninformative Textual Styles for AI-Generated Peer Review Detection [60.52240468810558]
  We introduce CoCoNUTS, a content-oriented benchmark built upon a fine-grained dataset of AI-generated peer reviews.<n>We also develop CoCoDet, an AI review detector via a multi-task learning framework, to achieve more accurate and robust detection of AI involvement in review content.
  arXiv  Detail & Related papers  (2025-08-28T06:03:11Z)
• Rethinking Code Review Workflows with LLM Assistance: An Empirical Study [2.9593087583214173]
  This paper combines an exploratory field study of current code review practices with a field experiment involving two variations of an LLM-assisted code review tool.<n>The study identifies key challenges in traditional code reviews, including frequent context switching and insufficient contextual information.<n>In the field experiment, we developed two prototype variations: one offering LLM-generated reviews upfront and the other enabling on-demand interaction.
  arXiv  Detail & Related papers  (2025-05-22T07:54:07Z)
• LazyReview A Dataset for Uncovering Lazy Thinking in NLP Peer Reviews [74.87393214734114]
  This work introduces LazyReview, a dataset of peer-review sentences annotated with fine-grained lazy thinking categories.<n>Large Language Models (LLMs) struggle to detect these instances in a zero-shot setting.<n> instruction-based fine-tuning on our dataset significantly boosts performance by 10-20 performance points.
  arXiv  Detail & Related papers  (2025-04-15T10:07:33Z)
• Identifying Aspects in Peer Reviews [59.02879434536289]
  We develop a data-driven schema for deriving aspects from a corpus of peer reviews.<n>We introduce a dataset of peer reviews augmented with aspects and show how it can be used for community-level review analysis.
  arXiv  Detail & Related papers  (2025-04-09T14:14:42Z)
• Harnessing Large Language Models for Curated Code Reviews [2.5944208050492183]
  In code review, generating structured and relevant comments is crucial for identifying code issues and facilitating accurate code changes.<n>Existing code review datasets are often noisy and unrefined, posing limitations to the learning potential of AI models.<n>We propose a curation pipeline designed to enhance the quality of the largest publicly available code review dataset.
  arXiv  Detail & Related papers  (2025-02-05T18:15:09Z)
• Too Noisy To Learn: Enhancing Data Quality for Code Review Comment Generation [2.990411348977783]
  Open-source datasets are used to train neural models for automated code review tasks.<n>These datasets contain a significant amount of noisy comments that persist despite cleaning methods.<n>We propose a novel approach using large language models (LLMs) to further clean these datasets.
  arXiv  Detail & Related papers  (2025-02-04T22:48:58Z)
• Deep Assessment of Code Review Generation Approaches: Beyond Lexical Similarity [27.92468098611616]
  We propose two novel semantic-based approaches for assessing code reviews.<n>The first approach involves converting both the generated review and its reference into digital vectors using a deep learning model.<n>The second approach generates a prompt based on the generated review and its reference, submits this prompt to ChatGPT, and requests ChatGPT to rate the generated review.
  arXiv  Detail & Related papers  (2025-01-09T11:52:32Z)
• Deep Just-In-Time Inconsistency Detection Between Comments and Source Code [51.00904399653609]
  In this paper, we aim to detect whether a comment becomes inconsistent as a result of changes to the corresponding body of code. We develop a deep-learning approach that learns to correlate a comment with code changes. We show the usefulness of our approach by combining it with a comment update model to build a more comprehensive automatic comment maintenance system.
  arXiv  Detail & Related papers  (2020-10-04T16:49:28Z)
• Unsupervised Opinion Summarization with Noising and Denoising [85.49169453434554]
  We create a synthetic dataset from a corpus of user reviews by sampling a review, pretending it is a summary, and generating noisy versions thereof. At test time, the model accepts genuine reviews and generates a summary containing salient opinions, treating those that do not reach consensus as noise.
  arXiv  Detail & Related papers  (2020-04-21T16:54:57Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.