Diffs vs. Whole Files: An Empirical Comparison of Iterative Edit-Based and Direct Generation for Flutter/Dart Code Models

  • 类型：arxiv
  • 标识：2609.05779
  • 链接：https://arxiv.org/abs/2609.05779
  • 主分类：evaluation
  • 形态：benchmark
  • 被引：0
  • 被引来源：Semantic Scholar + OpenAlex
  • S2被引：0
  • OpenAlex被引：0
  • 影响力被引：0
  • TLDR：A single, architecture-independent mechanism behind the conditions where diff-based generation does win is identified: it is competitive on short, spatially localized edits, and its category-level wins concentrate in exactly the two task categories - refactoring and error-handling/edge-case fixes - with the lowest mean edit-step count in the authors' dataset.
  • OpenAlex ID：W7212025276
  • OpenAlex DOI：10.48550/arxiv.2609.05779
  • DOI：10.48550/arxiv.2609.05779
  • DOI来源：OpenAlex
  • 开放获取：green
  • 开放获取链接：https://arxiv.org/pdf/2609.05779
  • OpenAlex更新：2026-09-12
  • 待LLM分类：否
  • 成熟度：research
  • 场景：code-generation、empirical-comparison、edit-based
  • 标题中文：Diff 还是整文件：Flutter/Dart 代码模型中迭代编辑式生成与直接生成的实证比较
  • TLDR中文：识别出 diff 生成能够胜出的一种与架构无关的统一机制：它在短小且空间局部化的编辑上具有竞争力；其类别级优势恰好集中在作者数据集中平均编辑步数最低的两类任务——重构与错误处理/边界用例修复。
  • 来源文件：
  • /inbox/tom/_candidates/2026-09-10-agent-rag-longcontext-candidates.json
  • [S2 enrich]
  • [OpenAlex backfill]