karma/crab-webapp
1
0
Fork 0
mirror of https://github.com/karma-riuk/crab-webapp.git synced 2025-07-04 22:08:12 +02:00
Code Issues Packages Projects Releases Wiki Activity

made a better about modal content

Browse Source
This commit is contained in:
Karma Riuk
2025-06-16 17:23:04 +02:00
parent 4950264fb3
commit 57b4900e5a
2 changed files with 40 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
public/css/style.css
View File

@ -47,7 +47,7 @@ header .info-btn {
padding: 0;
}
fieldset .info-btn {
fieldset .info-btn, .modal-overlay .info-btn{
padding: revert;
}

53
public/index.html
View File

@ -58,7 +58,7 @@
<br /><br />
<div style="display: flex; align-items: center; gap: 0.5em">
<button id="upload-btn">Upload JSON</button>
<div id="upload-status" class="hidden" style="color: green;"> hello world </div>
<div id="upload-status" class="hidden" style="color: green;"></div>
</div>
</fieldset>
@ -127,20 +127,45 @@
<template id="about">
<h2>About this project</h2>
<div>
<p>CRAB (Code Review Automation Benchmark) is a research-driven platform designed to evaluate deep
learning models for code review tasks. Developed as part of a master's thesis at the Università
della Svizzera italiana, CRAB provides a high-quality, curated benchmark dataset of Java code review
triplets: submitted code, reviewer comment, and revised code. Each instance is manually validated to
ensure that reviewer comments directly address code issues and that the revised code implements the
feedback accurately. </p>
<p>The platform supports two core tasks: generating human-like review comments and refining code based
on those comments. It also accounts for paraphrased feedback and alternative valid code revisions,
offering a more realistic and robust evaluation. CRAB addresses the shortcomings of existing
datasets by eliminating noise and ensuring functional correctness through testing. Researchers can
upload model predictions to receive standardized evaluations, making CRAB an essential tool for
advancing automated code review technologies.</p>
<p>
This project introduces <strong>CRAB (Code Review Automated Benchmark)</strong>, a high-quality
benchmark designed to evaluate deep learning-based code review automation tools. It focuses on two
key tasks:
</p>
<ul>
<li><strong>Comment Generation</strong>: Generating natural language review comments that identify
issues and suggest improvements for a given piece of code.</li>
<li><strong>Code Refinement</strong>: Producing revised code that correctly implements the
suggestions from a review comment.</li>
</ul>
<p>
The dataset consists of
carefully curated triplets
<code>&lt;submitted_code, reviewer_comment, revised_code&gt;</code>—ensuring each comment is
actionable and each revision implements the suggested change. This eliminates noise common in
previous datasets and supports reliable, meaningful evaluation.
</p>
<p>
To support model benchmarking, we also provide a web-based evaluation platform (the website on which
you are reading this description) that allows researchers to download the dataset, submit their
predictions, and assess model performance across both tasks.
</p>
<p>
You can explore the source code for each component here:
</p>
<ul>
<li><a href="https://github.com/karma-riuk/crab" target="_blank">Dataset Construction Repository</a>
</li>
<li><a href="https://github.com/karma-riuk/crab-webapp" target="_blank">Web App Repository</a></li>
</ul>
<p>
This website lets you evaluate code review models against the CRAB benchmark. You can download input
files for either the comment generation or code refinement task, upload your model’s predictions,
and view the results once processing is complete. Each section includes a help icon
<button class='info-btn'><i class="fa fa-info"></i></button> that provides more detailed
instructions and file format guidelines.
</p>
</div>
</template>