9 papers
From Noisy Traces to Root Causes: Structural Trajectory Analysis and Causal Extraction for Agent Optimization
Ying Chang, Jiahang Xu, Xuan Feng +3
The optimization of long-horizon agents increasingly relies on reflection-based mechanisms, where a large language model (LLM) acts as an optimizer to diagnose agent failures and i…
Pull Requests as a Training Signal for Repo-Level Code Editing
Qinglin Zhu, Tianyu Chen, Shuai Lu +8
Repository-level code editing requires models to understand complex dependencies and execute precise multi-file modifications across a large codebase. While recent gains on SWE-ben…
From Patches to Trajectories: Privileged Process Supervision for Software-Engineering Agents
Murong Ma, Tianyu Chen, Yun Lin +7
Supervised fine-tuning (SFT) on long teacher trajectories is the dominant way to instill investigation and reasoning in open software-engineering (SWE) agents. Since every retained…
Reducing the Costs of Proof Synthesis on Rust Systems by Scaling Up a Seed Training Set
Nongyu Di, Tianyu Chen, Shan Lu +6
Large Language Models (LLMs) are widely used for code generation. However, the correctness of code generated by LLMs remains a concern. A potential remedy to this concern is to hav…
Data Mixing Agent: Learning to Re-weight Domains for Continual Pre-training
Kailai Yang, Xiao Liu, Lei Ji +6
Continual pre-training on small-scale task-specific data is an effective method for improving large language models in new target fields, yet it risks catastrophic forgetting of th…
Automated Proof Generation for Rust Code via Self-Evolution
Tianyu Chen, Shuai Lu, Shan Lu +11
Ensuring correctness is crucial for code generation. Formal verification offers a definitive assurance of correctness, but demands substantial human effort in proof construction an…