跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 25 天前AI 評分38

CodeSpecWeave 如何透過互相依存的規格擴展長期編碼任務

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

AI 導讀

CodeSpecWeave 透過合成帶有互相依存規格及經驗證測試的軟件開發任務,訓練編碼智能體,提升其在長期基準測試上的表現。研究團隊以此流程建立 4,670 項訓練任務,涵蓋 2,865 個程式碼庫;在這些任務上進行強化學習,模型表現顯著優於基礎模型及現有基準方法。

正文

Published on Oct 29

·

Submitted by

yi

on Nov 2

Authors:

,

,

,

,

,

,

Abstract

CodeSpecWeave synthesizes complex, interdependent software development tasks with verified tests to train coding agents, improving performance on long-horizon benchmarks.

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.

View arXiv page View PDF Project page Add to collection

Get this paper in your agent:

hf papers read 2610.40256

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2610.40256 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2610.40256 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2610.40256 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co