# Predicting Acceptance and Review Effort in Human and Agent Pull Requests

arXiv: https://arxiv.org/abs/2607.12057 · HTML: https://mlsanity.com/paper/2607.12057/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Jul 2026.

## Built on by (most cited first, 1 of 1)

- [Audit the Scaffold, Not the Checkpoint: A Stationarity Dichotomy for Recursive Self-Improvement in Agentic Coding](https://mlsanity.com/paper/2609.34924/) (2026)

## Abstract (arXiv)

Pull requests (PRs) are a central mechanism for reviewing and integrating code changes in modern software repositories. As AI coding agents begin to submit more code changes alongside human developers, maintainers face a new challenge: deciding which PRs are likely to be accepted and which ones may require substantial review effort. This paper studies whether such outcomes can be estimated at the time a PR is opened, before reviewer discussion, CI feedback, or merge decisions are available. Using the AIDev dataset, we construct a leakage-aware prediction pipeline for human- and agent-authored PRs. The feature set is limited to submission-time information, including PR text characteristics, metadata, repository context, temporal signals, and lightweight diff statistics. We evaluate classical machine-learning models, including Logistic Regression, Random Forests, Gradient Boosting, Extra Trees, and MLPs, across pooled, human-only, agent-only, and balanced contributor views. Our results show that acceptance prediction is feasible from early signals: tree-based models achieve F1 scores above 0.95, with textual clarity and metadata among the most influential predictors. Review-effort prediction is more difficult. Comment counts and time-to-merge are only modestly explained by submission-time features, suggesting that reviewer availability, project workflow, and team-specific review practices play a major role. These findings indicate that early PR models can support triage and reviewer prioritization, but should be used as advisory tools rather than automated decision-makers.