---
title: "SWE-INTERACT：重新构想用户驱动的长时间编码会话的软件工程基准"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2606.30573v1"
permalink: "https://feed.astralor.com/items/itm_ebc400b5561017d1f699cbf0"
published: "2026-06-29T17:17:45.000Z"
score: 58
channel: "ai"
tags: ["GPT-5.5","Opus 4.8","SWE-INTERACT","多轮交互基准","用户模拟器","编码代理"]
---

# SWE-INTERACT：重新构想用户驱动的长时间编码会话的软件工程基准

## 速览

SWE-Interact是一个新的测试平台，用于评估编码代理在多轮交互、用户驱动的软件工程任务中的表现。与现有基准不同，它通过精心设计的用户模拟器从模糊或不完整的指令开始，逐步揭示需求，检查工作区，并提供针对性反馈，直到全部任务目标移交。实验表明，模型在单轮任务中的强性能无法可靠迁移到多轮用户驱动工作流：最佳模型解决约50%单轮任务，但在SWE-Interact中仅25%。Opus 4.8和GPT 5.5等最强模型在模糊指令下表现良好，但仍存在过度代理、遗忘需求等技术问题。该基准衡量了交互式目标发现与迭代细化这一正交的真实世界能力。

## 关键点

- 提出SWE-Interact基准，模拟用户驱动的多轮软件工程任务，从模糊需求开始逐步细化。
- 主流模型在单轮任务中表现优异，但在多轮交互任务中性能显著下降（从50%降至25%）。
- 最强模型如Opus 4.8和GPT 5.5能处理模糊指令，但存在过度代理、遗忘需求和代码错误。

## 全文

We introduce SWE-Interact, a new testbed for evaluating coding agents on multi-turn, interactive, user-driven software engineering tasks. Existing frontier SWE benchmarks typically provide complete requirements upfront and evaluate agents on autonomous implementation. In contrast, SWE-Interact places agents in a realistic developer workflow: a carefully designed user simulator starts with vague or incomplete instructions, progressively reveals requirements, inspects the agent's workspace, and provides targeted feedback, revisions, and new constraints until the full task goal has been handed off. Grounded in large-scale studies of real coding-agent interactions, this setup tests whether agents can discover user intent, adapt to evolving requirements, and build on their own prior work. Across a suite of frontier and open-weight models, we find that strong performance on single-turn SWE tasks does not reliably transfer to multi-turn, user-driven workflows: the best-performing models solve roughly 50% of single-turn baseline tasks but only 25% of the corresponding SWE-Interact tasks. The strongest models in our evaluation, including Opus 4.8 and GPT 5.5, start strong even in the face of vague initial instructions, persevere until all the requirements are surfaced by the user, integrate them better and write clean code. However, they still suffer from over-agentic coding, forgetting requirements and technical mistakes. Weaker models start poorly under ambiguity, give up early, forget or ignore instructions and rework their code more. Overall, SWE-Interact measures an orthogonal, real-world capability axis for frontier model development: interactive goal discovery and iterative refinement with a user in the loop.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_ebc400b5561017d1f699cbf0