You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Based on a systematic review of **201 papers and online resources**, this survey establishes a holistic theoretical framework for Issue Resolution in software engineering. We examine how **Large Language Models (LLMs)** are transforming the automation of GitHub issue resolution. Beyond the theoretical analysis, we have curated a comprehensive collection of datasets and model training resources, which are continuously synchronized with our GitHub repository and project documentation website.
33
+
Based on a systematic review of **203 papers and online resources**, this survey establishes a holistic theoretical framework for Issue Resolution in software engineering. We examine how **Large Language Models (LLMs)** are transforming the automation of GitHub issue resolution. Beyond the theoretical analysis, we have curated a comprehensive collection of datasets and model training resources, which are continuously synchronized with our GitHub repository and project documentation website.
34
34
35
35
## 📰 News
36
36
@@ -44,13 +44,13 @@ Based on a systematic review of **201 papers and online resources**, this survey
44
44
-**SWE-Adept**: SWE-Adept: An LLM-Based Agentic Framework for Deep Codebase Analysis and Structured Issue Resolution [](https://arxiv.org/abs/2603.01327)
-**SWE-CI**: SWE-CI: Evaluating Agent Capabilities in Maintaining Codebases via Continuous Integration [](https://arxiv.org/abs/2603.03823)[](https://github.com/SKYLENAGE-AI/SWE-CI)[](https://huggingface.co/datasets/skylenage/SWE-CI)
47
+
-**SWE-Skills-Bench**: SWE-Skills-Bench: Do Agent Skills Actually Help in Real-World Software Engineering? [](https://arxiv.org/abs/2603.15401)[](https://github.com/GeniusHTX/SWE-Skills-Bench)
48
+
-**OpenSWE**: daVinci-Env: Open SWE Environment Synthesis at Scale [](https://arxiv.org/abs/2603.13023)[](https://github.com/GAIR-NLP/OpenSWE)[](https://huggingface.co/datasets/GAIR/OpenSWE)
47
49
-**Closing the Loop**: Closing the Loop: Universal Repository Representation with RPG-Encoder [](https://arxiv.org/abs/2602.02084)[](https://github.com/microsoft/RPG-ZeroRepo)[](https://ayanami2003.github.io/RPG-Encoder/)
48
50
-**DockSmith**: DockSmith: Scaling Reliable Coding Environments via an Agentic Docker Builder [](https://arxiv.org/abs/2602.00592)[](https://huggingface.co/collections/8sj7df9k8m5x8/docksmith)
49
-
-**Rust-SWE-bench**: Evaluating and Improving Automated Repository-Level Rust Issue Resolution with LLM-based Agents [](https://arxiv.org/abs/2602.22764)[](https://github.com/GhabiX/Rust-SWE-Bench)
50
51
-**Scale-SWE**: Immersion in the GitHub Universe: Scaling Coding Agents to Mastery [](https://arxiv.org/abs/2602.09892)[](https://github.com/AweAI-Team/ScaleSWE)[](https://huggingface.co/collections/AweAI-Team/scale-swe)
51
52
-**SWE Context Bench**: SWE Context Bench: A Benchmark for Context Learning in Coding [](https://arxiv.org/pdf/2602.08316)
-**SWE-Bench Mobile**: SWE-Bench Mobile: Can Large Language Model Agents Develop Industry-Level Mobile Applications? [](https://arxiv.org/abs/2602.09540)[](https://swebenchmobile.com/)
54
54
-**SWE-Hub**: SWE-Hub: A Unified Production System for Scalable, Executable Software Engineering Tasks [](https://arxiv.org/abs/2603.00575)
55
55
-**SWE-Master**: SWE-Master: Unleashing the Potential of Software Engineering Agents via Post-Training [](https://arxiv.org/abs/2602.03411)[](https://github.com/RUCAIBox/SWE-Master)
56
56
-**SWE-MiniSandbox**: SWE-MiniSandbox: Container-Free Reinforcement Learning for Building Software Engineering Agents [](https://arxiv.org/abs/2602.11210v1)[](http://github.com/lblankl/SWE-MiniSandbox)
@@ -109,7 +109,7 @@ Based on a systematic review of **201 papers and online resources**, this survey
109
109
## 📚 Complete Paper List
110
110
111
111
112
-
> **Total: 201 works** across 14 categories
112
+
> **Total: 203 works** across 14 categories
113
113
114
114
115
115
### 📊 Evaluation Datasets
@@ -119,6 +119,7 @@ Based on a systematic review of **201 papers and online resources**, this survey
119
119
-`(2026-03)`**BeyondSWE**: BeyondSWE: Can Current Code Agent Survive Beyond Single-Repo Bug Fixing? [](https://arxiv.org/abs/2603.03194)[](https://aweai-team.github.io/BeyondSWE/)[](https://github.com/AweAI-Team/BeyondSWE)[](https://huggingface.co/datasets/AweAI-Team/BeyondSWE)
120
120
-`(2026-03)`**SWE-CI**: SWE-CI: Evaluating Agent Capabilities in Maintaining Codebases via Continuous Integration [](https://arxiv.org/abs/2603.03823)[](https://github.com/SKYLENAGE-AI/SWE-CI)[](https://huggingface.co/datasets/skylenage/SWE-CI)
-`(2026-03)`**SWE-Skills-Bench**: SWE-Skills-Bench: Do Agent Skills Actually Help in Real-World Software Engineering? [](https://arxiv.org/abs/2603.15401)[](https://github.com/GeniusHTX/SWE-Skills-Bench)
122
123
-`(2026-02)`**SWE Context Bench**: SWE Context Bench: A Benchmark for Context Learning in Coding [](https://arxiv.org/pdf/2602.08316)
-`(2026-02)`**Rust-SWE-bench**: Evaluating and Improving Automated Repository-Level Rust Issue Resolution with LLM-based Agents [](https://arxiv.org/abs/2602.22764)[](https://github.com/GhabiX/Rust-SWE-Bench)
@@ -148,6 +149,7 @@ Based on a systematic review of **201 papers and online resources**, this survey
148
149
149
150
*Datasets for training issue resolution agents*
150
151
152
+
-`(2026-03)`**OpenSWE**: daVinci-Env: Open SWE Environment Synthesis at Scale [](https://arxiv.org/abs/2603.13023)[](https://github.com/GAIR-NLP/OpenSWE)[](https://huggingface.co/datasets/GAIR/OpenSWE)
151
153
-`(2026-02)`**SWE-Universe**: SWE-Universe: Scale Real-World Verifiable Environments to Millions [](https://www.arxiv.org/abs/2602.02361)
152
154
-`(2026-02)`**SWE-rebench V2**: SWE-rebench V2: Language-Agnostic SWE Task Collection at Scale [](https://arxiv.org/abs/2602.23866)
153
155
-`(2026-02)`**Scale-SWE**: Immersion in the GitHub Universe: Scaling Coding Agents to Mastery [](https://arxiv.org/abs/2602.09892)[](https://github.com/AweAI-Team/ScaleSWE)[](https://huggingface.co/collections/AweAI-Team/scale-swe)
@@ -259,6 +261,7 @@ Based on a systematic review of **201 papers and online resources**, this survey
259
261
260
262
*Models trained via supervised learning*
261
263
264
+
-`(2026-03)`**OpenSWE**: daVinci-Env: Open SWE Environment Synthesis at Scale [](https://arxiv.org/abs/2603.13023)[](https://github.com/GAIR-NLP/OpenSWE)[](https://huggingface.co/datasets/GAIR/OpenSWE)
262
265
-`(2026-02)`**Scale-SWE**: Immersion in the GitHub Universe: Scaling Coding Agents to Mastery [](https://arxiv.org/abs/2602.09892)[](https://github.com/AweAI-Team/ScaleSWE)[](https://huggingface.co/collections/AweAI-Team/scale-swe)
263
266
-`(2026-01)`**SWE-Lego**: SWE-Lego: Pushing the Limits of Supervised Fine-tuning for Software Issue Resolving [](https://arxiv.org/abs/2601.01426)
264
267
-`(2026-01)`**SWE-Replay**: SWE-Replay: Efficient Test-Time Scaling for Software Engineering Agents [](https://arxiv.org/abs/2601.22129)
@@ -333,6 +336,7 @@ Based on a systematic review of **201 papers and online resources**, this survey
333
336
334
337
*Techniques for collecting training data*
335
338
339
+
-`(2026-03)`**OpenSWE**: daVinci-Env: Open SWE Environment Synthesis at Scale [](https://arxiv.org/abs/2603.13023)[](https://github.com/GAIR-NLP/OpenSWE)[](https://huggingface.co/datasets/GAIR/OpenSWE)
336
340
-`(2026-02)`**DockSmith**: DockSmith: Scaling Reliable Coding Environments via an Agentic Docker Builder [](https://arxiv.org/abs/2602.00592)[](https://huggingface.co/collections/8sj7df9k8m5x8/docksmith)
337
341
-`(2026-02)`**SWE-rebench V2**: SWE-rebench V2: Language-Agnostic SWE Task Collection at Scale [](https://arxiv.org/abs/2602.23866)
338
342
-`(2026-02)`**Scale-SWE**: Immersion in the GitHub Universe: Scaling Coding Agents to Mastery [](https://arxiv.org/abs/2602.09892)[](https://github.com/AweAI-Team/ScaleSWE)[](https://huggingface.co/collections/AweAI-Team/scale-swe)
0 commit comments