To improve large language models'ability to resolve real-world software issues, prior work has focused on constructing large-scale agent trajectory datasets and performing supervised fine-tuning (SFT) on successful trajectories. However, task success does not guarantee high-quality supervision: successful trajectories...
De-Wu Zheng, Rui Ye, Yanlan Wang et al.· 3 citations
PTA-IRT is proposed, a Privileged Trajectory-Aware Item Response Theory framework that fuses process and outcome signals and consistently outperforms prior IRT baselines on score and ranking recovery across four SWE benchmarks.
Ke-Feng Duan, De-Wu Zheng, Yan-Lin Wang et al.· 0 citations
ACToR identifies critical tokens during generation and triggers targeted retrieval on demand to provide repository context at these decisive positions, and designs a position-aware weighting method for dense retrievers to prioritize context that is more informative for generation.
Ke-Feng Duan, De-Wu Zheng, Yan-Lin Wang et al.· 0 citations
MCR-Bench is introduced, the first defect state-aware benchmark designed for realistic multi-round code review, and in-depth error analysis dissects the distinct drivers of false positives and false negatives, revealing critical weaknesses such as cross-round temporal misalignment and inadequate long-range memory.
De-Wu Zheng, Yan-Lin Wang, Xi-Wen Wang et al.· Proceedings of the ACM on So...· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
