StepCodeReasoner aligns code reasoning with verifiable stepwise execution traces via print anchors and bi-level GRPO reinforcement learning, reaching SOTA results on CRUXEval (91.1%) and LiveCodeBench (86.5%) for a 7B model.
Lingma swe-gpt: An open development-process-centric language model for automated software improvement
10 Pith papers cite this work, alongside 1 external citations. Polarity classification is still indexing.
10
Pith papers citing it
1
external citations · external index
citation-role summary
background 3
citation-polarity summary
roles
background 3polarities
background 3representative citing papers
ADI equips AI debugging agents with function-level interaction via a new execution trace structure, raising SWE-bench Verified resolution to 63.8% at $1.28 per task and delivering 6-18% gains when added to existing agents.
A purpose-built, staged LLM agent correctly sets up and executes software analysis tools on 33 of 35 benchmark tasks, outperforming general-purpose agent baselines by at least 17 percentage points.
SWE-RL uses RL on software evolution data to train LLMs achieving 41% on SWE-bench Verified with generalization to other reasoning tasks.
BoostAPR boosts automated program repair by training a sequence-level assessor and line-level credit allocator from execution outcomes, then applying them in PPO to reach 40.7% on SWE-bench Verified.
Agent-CoEvo is a multi-agent LLM framework that coevolves code patches and test patches to resolve repository-level issues, outperforming fixed-test baselines on SWE-bench Lite and SWT-bench Lite.
Empirical study of 3977 agent trajectories finds Python execution errors correlate with lower success rates on GitHub issues, flags challenging errors, and reports three confirmed bugs in the SWE-Bench platform.
This survey categorizes agentic environments for LLMs by eight attributes and domains, introduces symbolic and neural synthesis paradigms with evaluation, and outlines four agent evolution pathways plus three environment evolution paradigms.
Non-linear domain-scoped parallel LLM agents achieve higher micro F1 than linear exploration and some baselines for multi-file change localization on SWE-bench Pro ansible tasks.
AI agents restructure software from static pre-written code to dynamic LLM-driven systems, transferring decision complexity and expanding software engineering into Agentic Engineering.
citing papers explorer
-
StepCodeReasoner: Aligning Code Reasoning with Stepwise Execution Traces via Reinforcement Learning
StepCodeReasoner aligns code reasoning with verifiable stepwise execution traces via print anchors and bi-level GRPO reinforcement learning, reaching SOTA results on CRUXEval (91.1%) and LiveCodeBench (86.5%) for a 7B model.
-
Empowering Autonomous Debugging Agents with Efficient Dynamic Analysis
ADI equips AI debugging agents with function-level interaction via a new execution trace structure, raising SWE-bench Verified resolution to 63.8% at $1.28 per task and delivering 6-18% gains when added to existing agents.
-
Evaluating LLM Agents on Automated Software Analysis Tasks
A purpose-built, staged LLM agent correctly sets up and executes software analysis tools on 33 of 35 benchmark tasks, outperforming general-purpose agent baselines by at least 17 percentage points.
-
SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution
SWE-RL uses RL on software evolution data to train LLMs achieving 41% on SWE-bench Verified with generalization to other reasoning tasks.
-
BoostAPR: Boosting Automated Program Repair via Execution-Grounded Reinforcement Learning with Dual Reward Models
BoostAPR boosts automated program repair by training a sequence-level assessor and line-level credit allocator from execution outcomes, then applying them in PPO to reach 40.7% on SWE-bench Verified.
-
Beyond Fixed Tests: Repository-Level Issue Resolution as Coevolution of Code and Behavioral Constraints
Agent-CoEvo is a multi-agent LLM framework that coevolves code patches and test patches to resolve repository-level issues, outperforming fixed-test baselines on SWE-bench Lite and SWT-bench Lite.
-
Beyond Final Code: A Process-Oriented Error Analysis of Software Development Agents in Real-World GitHub Scenarios
Empirical study of 3977 agent trajectories finds Python execution errors correlate with lower success rates on GitHub issues, flags challenging errors, and reports three confirmed bugs in the SWE-Bench platform.
-
Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Application
This survey categorizes agentic environments for LLMs by eight attributes and domains, introduces symbolic and neural synthesis paradigms with evaluation, and outlines four agent evolution pathways plus three environment evolution paradigms.
-
Exploration Structure in LLM Agents for Multi-File Change Localization
Non-linear domain-scoped parallel LLM agents achieve higher micro F1 than linear exploration and some baselines for multi-file change localization on SWE-bench Pro ansible tasks.
-
Agentic Software: How AI Agents Are Restructuring the Software Paradigm
AI agents restructure software from static pre-written code to dynamic LLM-driven systems, transferring decision complexity and expanding software engineering into Agentic Engineering.