MEGA Hub

When Requirements Interact: Scaling Long-Horizon Coding via Interdependent Specifications

Authors

Do you know Yishuo Cai?You can claim authorship or link another user.Do you know Minzheng Wang?You can claim authorship or link another user.Do you know Liang Zhao?You can claim authorship or link another user.Do you know Linghao Zhang?You can claim authorship or link another user.Do you know Lei Li?You can claim authorship or link another user.Do you know Shicheng Li?You can claim authorship or link another user.Do you know Xu Sun?You can claim authorship or link another user.Do you know Fuli Luo?You can claim authorship or link another user.

Abstract

To succeed in real-world long-horizon software development, coding agents must satisfy multi-dimensional task specifications, such as interface compatibility and error handling, and do so without breaking existing dependencies. However, current works lack specification in both quantity and diversity, often emphasizing primary goals while ignoring complex constraints. Synthesizing tasks to bridge this gap presents two challenges: naively adding specifications risks reducing tasks to trivial checklists rather than compelling joint reasoning, and verifying these subtle interactions risks overlooking hidden flaws or rejecting valid alternatives. To address these challenges, we introduce CodeSpecWeave, a framework for synthesizing tasks with rich, interdependent specifications. Grounded in real developer seeds, a synthesis agent iteratively develops both coupled task requirements and executable tests, while a review agent audits solution trajectories from multiple models to guarantee test reliability. Using this pipeline, we construct 4,670 training tasks across 2,865 repositories. Applying Reinforcement learning on these tasks significantly improves performance over the base model on DeepSWE-v1.1, largely outperforming existing baselines. Furthermore, we curate CodeSpecWeave-Bench, a rigorous benchmark comprising 100 tasks and 1,044 specifications, which clearly separates frontier models and reveals significant headroom in handling interdependent engineering constraints.

Community

00