100 REAL TASKS · 6 DOMAINS · HONEST RUN 2026-09-11
WebAgent Failure Atlas
中文真实场景 AI 浏览器 Agent 失败图鉴 — 同一套 harness 真抓每一步 HTTP 与快照 hash,失败即数据。镜像英文基准 WebArena / Mind2Web / VisualWebArena / WebVoyager / OSWorld 的方法学,换成中文真网。
总体通过率 Overall
68.0% 68/100
95% Wilson CI 58.3%–76.3%
最硬墙 / 最稳
政务 29% · 电商 100%
ecom 17/17 · gov 5/17 · χ² p=0.0007
失败 Top2
F6×14 · F3×8
中文表单语义 · 动态渲染未等待
干预实证
1/32 3.1%
5s 退避重试仅救回 search-01 (F3)
🇨🇳 中文

研究亮点

  • 100 个中文真实任务:搜索 17 / 电商 17 / 表单 17 / 政务 17 / 地图 16 / 媒体 16
  • 总体成功率 68.0%(95% Wilson CI 58.3%–76.3%)
  • 政务最硬墙(5/17,29%)vs 电商最稳(17/17,100%)
  • 失败 Top2:中文表单语义 F6×14、动态渲染未等待 F3×8
  • 重试只救回 3.1%(1/32),证明软墙需结构性修复
  • 卡方检验 χ²=21.216,p=0.0007,领域差异极显著
🇺🇸 English

Key Findings

  • 100 real Chinese tasks: search 17 / ecom 17 / form 17 / gov 17 / map 16 / media 16
  • Overall 68.0% (95% Wilson CI 58.3%–76.3%)
  • Gov = hard wall (5/17, 29%) vs Ecom = trivial (17/17, 100%)
  • Top failures: F6 Chinese form semantics ×14, F3 dynamic rendering ×8
  • Retry rescued only 3.1% (1/32) — soft walls need structural fixes
  • Chi-square χ²=21.216, p=0.0007 — domain matters significantly
分域表现 / By Domain
Domain 领域npass 通过rate 通过率95% CI
🛒 ecom 电商1717100.0%81.6%–100.0%
🔍 search 搜索171376.5%52.7%–90.5%
🗺️ map 地图161275.0%50.5%–89.8%
📰 media 媒体161168.8%44.4%–85.8%
📝 form 表单171058.8%36.0%–78.4%
🏛️ gov 政务17529.4%13.3%–53.1%
失败分类 / Failure Taxonomy (7 classes, 5 observed)
Classn中文含义English Description
F614中文表单语义误解Chinese form semantics
F38动态加载未等待Dynamic rendering not waited
F14元素定位失败Element locator failure
F54验证码/反爬阻断Captcha / anti-scrape block
F42登录墙阻断Login wall block
F20视觉 grounding 错误Visual grounding error
F70多步遗忘/偏离Multi-step drift
🇨🇳 给 Builder 的 5 条修复杠杆
  1. 先等后断言:动态页固定等待 + 内容长度门限(<20KB 判 JS 壳),能吃掉大半 F3
  2. 备选选择器:同一语义 ≥3 种定位(文本/role/URL),一个不行换一个
  3. 登录墙早退:命中关键词即停并标记 F4,不浪费步数
  4. 反爬退避:403/429 指纹退避 + 降级镜像,保证轨迹不断档
  5. 中文表单专项:label→input 关联检查,无 label 直接告警 F6
🇺🇸 5 Fix Levers for Builders
  1. Wait before assert: fixed wait + <20KB JS-shell gate kills most F3
  2. Fallback selectors: ≥3 per semantic (text/role/URL), rotate on failure
  3. Login wall early exit: tag F4 immediately, save steps
  4. Anti-scrape backoff: 403/429 fingerprint + mirror fallback
  5. Chinese form check: label→input association, alert on missing label
一键复现 / One-click Reproduce
# One-click verify (秒级)
python reproduce.py
# → n=100 pass=68 rate=0.680 CI=[0.5834,0.7633]  retry 1/32

# Full rebuild (~2min)
bash verify.sh
# → stats + 9 figs + papers + dashboard
Single source of truth: stats.json → papers / dashboard / landing page all injected from this one file. Numbers are identical across all 4 channels.
单一事实源:stats.json → 论文/仪表盘/落地页均由此注入,四渠道数字完全一致。
9 张图 / 9 Figures
诚实边界 / Honest Limits: 中文:规则执行器 = 自动化下界(真实 LLM Agent 只会更高);n=100 CI 仍 ±9%;DOM 漂移 2026-09-11;观察性关联非因果;22 参考文献逐条 HTTP 200。AI 使用声明:代码/分析/文稿有 AI 辅助,科学判断与责任由人类作者承担。
English: Rule-based executor = lower bound (real LLM agents score higher); n=100 CI ±9%; DOM drift 2026-09-11; observational not causal; 22 refs HTTP-verified. AI disclosure: code/analysis/drafting had AI assistance; scientific judgments are human authors'.
方法学致谢 / Methodology Acknowledgments
WebArena (2307.13854) · Mind2Web (2306.06070) · VisualWebArena (2401.13649) · WebVoyager (2401.13919) · OSWorld (2404.07972) — 5 篇标杆奠定方法学,本研究镜像其纪律换成中文真网。
These 5 landmarks established the science of web agent evaluation — we port their discipline to the real Chinese web.
引用 / Cite
@dataset{atlas2026webagent,
  title={WebAgent Failure Atlas: 100 Real Chinese Tasks},
  year={2026},
  url={https://github.com/kevindurant735rocket-creator/webagent-failure-atlas}
}