Update README.md
Browse files
README.md
CHANGED
@@ -9,11 +9,23 @@ pinned: false
|
|
9 |
Model/Data associated with Paper:
|
10 |
|
11 |
|
12 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
|
14 |
-
[Jiayi Pan*](https://www.jiayipan.com/), [Xingyao Wang*](https://xwang.dev/), [Graham Neubig](https://www.phontron.com/), [Navdeep Jaitly](https://www.cs.toronto.edu/~ndjaitly/), [Ji Heng](https://blender.cs.illinois.edu/hengji.html), [Alane Suhr^](https://www.alanesuhr.com/), [Yizhe Zhang^](https://dreasysnail.github.io/)
|
15 |
-
|
16 |
-
UC Berkeley, UIUC, CMU, Apple | *, ^ denotes equal contribution
|
17 |
|
18 |
We present SWE-Gym, the first environment for training real-world software engineering agents. We use it to train strong LM agents that achieve state-of-the-art open results on SWE-Bench, with early, promising scaling characteristics as we increase training and inference-time compute.
|
19 |
|
|
|
9 |
Model/Data associated with Paper:
|
10 |
|
11 |
|
12 |
+
<h1 align="center"> Training Software Engineering Agents and Verifiers with SWE-Gym </h1>
|
13 |
+
|
14 |
+
<p align="center">
|
15 |
+
<a href="https://www.jiayipan.com/" style="text-decoration: none;">Jiayi Pan<sup>*,1</sup></a>,
|
16 |
+
<a href="https://xwang.dev/" style="text-decoration: none;">Xingyao Wang<sup>*,2</sup></a>,
|
17 |
+
<a href="https://www.phontron.com/" style="text-decoration: none;">Graham Neubig<sup>3</sup></a>,
|
18 |
+
<a href="https://www.cs.toronto.edu/~ndjaitly/" style="text-decoration: none;">Navdeep Jaitly<sup>4</sup></a>,
|
19 |
+
<a href="https://blender.cs.illinois.edu/hengji.html" style="text-decoration: none;">Ji Heng<sup>2</sup></a>,
|
20 |
+
<a href="https://www.alanesuhr.com/" style="text-decoration: none;">Alane Suhr<sup>^,1</sup></a>,
|
21 |
+
<a href="https://dreasysnail.github.io/" style="text-decoration: none;">Yizhe Zhang<sup>^,4</sup></a>
|
22 |
+
</p>
|
23 |
+
|
24 |
+
<p align="center">
|
25 |
+
<sup>1</sup>UC Berkeley, <sup>2</sup>UIUC, <sup>3</sup>CMU, <sup>4</sup>Apple </br>
|
26 |
+
<sub><sup>*</sup>Equal contribution, <sup>^</sup>Equal supervision</sub>
|
27 |
+
</p>
|
28 |
|
|
|
|
|
|
|
29 |
|
30 |
We present SWE-Gym, the first environment for training real-world software engineering agents. We use it to train strong LM agents that achieve state-of-the-art open results on SWE-Bench, with early, promising scaling characteristics as we increase training and inference-time compute.
|
31 |
|