发表机构
NTU Singapore（新加坡南洋理工大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
PaperCompiler是一种论文到代码生成框架，通过编译论文相关证据生成仓库级规范，在Paper2CodeBench上实现了优于基线的性能，提升了代码保真度并减少了高严重性批评。
AI 中文摘要
将研究论文忠实地转化为仓库级实现仍然具有挑战性，因为论文通常以高层级描述方法，隐含了实现假设，且要求生成的仓库保留方法逻辑、评估协议和跨文件一致性。尽管论文到代码智能体最近取得了进展，其中间输出常以自由格式的计划或摘要呈现，下游编码智能体可能忽略、重新解释或压缩这些内容，导致算法简化和仓库结构不一致。为应对这些挑战，我们提出PaperCompiler，一个论文到代码生成框架，它将基于论文的证据编译为显式的仓库级实现规范。PaperCompiler锚定与实现相关的证据，同时保留来源出处，并区分论文支持、推断、外部委托和未解决的信息。生成的规范编码了非降级要求、所有权分配、跨文件依赖和文件级约束。仓库生成在这些编译后的规范下进行，同时保留了论文未固定的本地工程选择的灵活性。PaperCompiler在Paper2CodeBench上优于强基线，在基于参考的保真度上实现了13.8%的相对提升（从3.64提升至4.15），并降低了高严重性评估者批评（从13.2%降至6.1%）。
英文摘要
Faithfully translating research papers into repository-level implementations remains challenging because papers often describe methods at a high level, leave implementation assumptions implicit, and require generated repositories to preserve method logic, evaluation protocols, and cross-file consistency. Despite recent advances in paper-to-code agents, their intermediate outputs are often presented as free-form plans or summaries that downstream coding agents may ignore, reinterpret, or compress, leading to algorithmic simplification and inconsistent repository structure. To address these challenges, we introduce PaperCompiler, a paper-to-code generation framework that compiles paper-grounded evidence into explicit repository-level implementation specifications. PaperCompiler grounds implementation-relevant evidence while preserving source provenance and distinguishing paper-supported, inferred, externally delegated, and unresolved information. The resulting specifications encode non-degradation requirements, ownership assignments, cross-file dependencies, and file-level constraints. Repository generation proceeds under these compiled specifications while retaining flexibility over local engineering choices not fixed by the paper. PaperCompiler outperforms strong baselines on Paper2CodeBench, achieving a 13.8% relative improvement in reference-based fidelity (from 3.64 to 4.15) and reducing high-severity evaluator critiques (from 13.2% to 6.1%).
Comments9 pages
