新的SWE-DIVE框架评估多轮编码代理

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-23 04:00

研究人员推出了SWE-DIVE，一个旨在评估动态、多轮交互中编码代理的新型框架。与使用单一、定义明确的提示的现有基准不同，SWE-DIVE包含一个能够模仿各种行为模式并迭代改进任务的用户模拟器。对话验证器评估模拟器的输出，以确保它们符合初始要求并引导代理完成任务。这个在Harbor环境中协调的综合系统旨在更好地反映现实世界中编码代理的使用情况。 AI

影响该框架可能导致对编码代理进行更现实的评估，从而提高它们处理复杂、交互式任务的能力。

排序理由该集群包含一篇介绍AI编码代理新评估框架的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 4 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇介绍AI编码代理新评估框架的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, product; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Lorenz Brehme, Benjamin Steenhoek, Ruth Breu · 2026-10-23 04:00

SWE-DIVE: Multi-Turn Coding Agent Evaluation with Dynamic User Simulation

arXiv:2610.30047v1 Announce Type: cross Abstract: Coding agents are typically evaluated using one-shot, well-formulated prompts. Benchmarks such as SWE-bench use existing GitHub issues as input for a coding agent and evaluate whether the agent successfully solves the task. Howeve…

本文如何被排名

报道来源 [1]

SWE-DIVE: Multi-Turn Coding Agent Evaluation with Dynamic User Simulation

相关实体

相关话题