Skip to content

part-0001 摘要与引言:一年期电商运营基准

摘要

Abstract 立论:LLM agent 越来越多地被期望处理跨数千步的任务,需要做序列决策、从过往 经验适应、并在初始指令之后长期保持连贯。作者把长 horizon agentic 任务按 RL 的 episodic(有终态)continuing(无终态) 二分,指出电商经营是 continuing 场景的 自然载体——高赌注、不可逆决策、复利式财务健康。现状基准各缺一角:Vending-Bench 系列模拟 多轮供应商谈判但用 mock 数据 + prompt 模拟 → 供应商反馈高度随机、可被操纵;YC-Bench 跑确定 性创业年但无供应商采购;MerchantBench / RetailBench 有真实商品与开放代码却放弃谈判与对抗 动态。论文提出 E-Commerce Bench:首个把确定性供应商谈判与客户销售建模统一进真实世界 continuing 经营的开源基准——365 天、最多 4 店并行、6,886 商品与 576 供应商(由淘宝/天猫 日志校准)、全年促销/灾害/供应链冲击日历;基准 harness 自动管理上下文窗口并提供持久记忆。 双确定性保证可复现:客户购买/退货走固定需求模型(唯一随机性来自 episode seed); 供应商侧确定性谈判内核固定每个定价/让步/接受决定,LLM 只渲染对话(约 1/4 供应商是欺诈方)。 评估 18 个前沿与开源模型,用主分数 + 六维指标而非单一总资产。

三大贡献:① 首个把确定性对手谈判、动态事件、并发多店管理融进一年 horizon 的开源基准; ② 双边确定性经济(需求模型 + 谈判内核,LLM 只做渲染)——内核把 TERMS-Bench 的对手设计 扩展到重复、跨供应商讨价还价;③ 18 模型多维评测——单个总资产数字不再能藏住弱点,最好 的赚钱者既不是最佳谈判者也不是最高效运营者。

§1 Introduction 简述背景:LLM 需在长轨迹上用外部工具规划与行动;episodic 基准(GitHub issue、真实电脑操作、专家评分知识工作)与 continuing 基准(Qian et al. 2025 交易、Chen et al. 2026)的区别,以及 business operation 作为 continuing 场景的恰当性;然后预告论文结构 (§2 相关工作 → §3 基准细节 → §4 结果与错误分析 → §5 结论)。

要点归纳

  • 论文坐标是「continuing 任务评测缺失」:现有基准要么偏离真实市场动态,要么靠纯 LLM 对手 → 抽样噪声毁掉可信度与可复现性
  • 双确定性是全文立身之本:结果差异「entirely belongs to the agent policy rather than to chance」
  • 开篇即埋两个贯穿全文的梗:欺诈约 1/4 供应商;「最能赚钱 ≠ 最能谈判/最高效」

术语 / 概念

  • episodic / continuing task(有终态 / 无终态任务) — RL 任务二分(Sutton & Barto 2018); 本文用其组织长 horizon agentic 基准谱系
  • deterministic negotiation kernel(确定性谈判内核) — 供应商侧固定策略决定价格/让步/ 接受,LLM 只渲染对话(详见 part-0003 / part-0009)
  • asset multiplier(资产倍数) — 期末总资产 ÷ ¥100,000 初始本金,主分数
  • Taobao & Tmall — 数据出处平台:6,886 商品、576 供应商、全年日历均以其日志校准

原句摘录

Business operations provide a natural continuing scenario with high-stakes, irreversible decisions, where compounding financial health tests sustained agent competence.

No sampling noise reaches the customers or the counterpart's pricing and accept/reject decisions, so a difference in outcome entirely belongs to the agent policy rather than to chance. Only the LLM renderer's wording varies between runs (§3.5.2).

GPT-5.6 Sol grows its capital roughly 14-fold, yet ranks 16th of 18 on fraud avoidance and earns less from a tool call than Fable5 does.

Reading six dimensions beside that total turns a leaderboard into a capability profile, and the profiles come out sharply divergent, the best earner being neither the best negotiator nor the efficient operator.

疑问 / 待查

  • 摘要只报 GPT-5.6 Sol / Claude Opus 4.7 / Qwen3.8-Max-Preview 三家的数字,其余 15 家的分布 留到 §4(part-0004)
  • 「the first open-source benchmark」的 first 主张依赖相关工作表(Table 1)的排他性,见 part-0002
Backlinks (1)
flowchart LR
  n0["ch-0001 凶案、侦探与母亲的信(第 2–163 章)"]
  n1["ch-0002 坦白、出逃与和解(第 163–233 章)"]
  n2["The Curious Incident of the Dog in the Night-Time 人物"]
  n3["The Curious Incident of the Dog in the Night-Time(整理完成)"]
  n4["The Curious Incident of the Dog in the Night-Time 笔记"]
  n5["The Curious Incident of the Dog in the Night-Time 故事线"]
  n6["E-Commerce Bench(整理完成)"]
  n7["E-Commerce Bench 笔记"]
  n8["part-0001 摘要与引言:一年期电商运营基准(Abstract + §1)"]
  n9["part-0002 相关工作与定位:continuing 任务谱系(Table 1 + §2)"]
  n10["part-0003 基准设计:四层架构与确定性经济(§3 主体)"]
  n11["part-0004 实验结果:18 模型的多维画像(Table 2 + §4)"]
  n12["part-0005 结论与参考文献(§5 + References)"]
  n13["part-0006 工具集与 agent harness:18 工具与上下文管理(Table 3 + Appendix A)"]
  n14["part-0007 数据层:类目、店型、日历与供应商(Appendix B)"]
  n15["part-0008 经济引擎:13 步结算与需求/退货/成本公式(Table 7 + Appendix C)"]
  n16["part-0009 确定性谈判内核:决策函数与骗局目录(Appendix D)"]
  n17["part-0010 评测指标定义:六个维度与统计口径(Appendix E)"]
  n18["part-0011 每维度结果导览与模型/实验设置(Table 10 + Appendix F)"]
  n19["part-0012 失败案例研究:骗局实录与一次破产(Appendix G)"]
  n20["part-0013 失败模式规则表与提示词设计(Table 18 + Appendix H)"]
  n21["The End of Software Engineering(整理完成)"]
  n22["The End of Software Engineering 笔记"]
  n23["part-0001 摘要与引言:范式重构的宣告(Abstract + §1 前段)"]
  n24["part-0002 第一性原理:传统软件与 Agent 系统的形式化模型(§1 后段 + §2)"]
  n25["part-0003 三代交付史与 AI→Software→Result 的失败(§3.1–3.2)"]
  n26["part-0004 Agent→Result 与 Agentic Engineering 学科(§3.3 + §4)"]
  n27["part-0005 实证证据与 EvoClaw 落差(§5)"]
  n28["part-0006 四阶段演进路线图(前段:表 3 + Stage I–III)"]
  n29["part-0007 路线图后段与建议:实践者与研究者(§6.3 后段 + §6.4 + §7.1–7.2)"]
  n30["part-0008 组织建议、结论与参考文献(§7.3 + §8)"]
  n31["ch-0001 Introduction(为什么做数据可视化)"]
  n32["ch-0002 选工具讲你的数据故事(Ch 1)"]
  n33["ch-0003 强化电子表格技能(Ch 2)"]
  n34["ch-0004 找到并质询你的数据(Ch 3)"]
  n35["ch-0005 清洗脏数据(Ch 4)"]
  n36["ch-0006 做有意义的比较(Ch 5)"]
  n37["ch-0007 图表化你的数据(Ch 6)"]
  n38["ch-0008 地图化你的数据(Ch 7)"]
  n39["ch-0009 表格化你的数据(Ch 8)"]
  n40["ch-0010 嵌入网页(Ch 9)"]
  n41["ch-0011 用 GitHub 编辑与托管代码(Ch 10)"]
  n42["ch-0012 Chart.js 与 Highcharts 模板(Ch 11)"]
  n43["ch-0013 Leaflet 地图模板(Ch 12)"]
  n44["ch-0014 转换你的地图数据(Ch 13)"]
  n45["ch-0015 识别谎言、减少偏差(Ch 14)"]
  n46["ch-0016 讲述并展示你的数据故事(Ch 15)"]
  n47["ch-0017 附录 A 排查常见问题"]
  n48["Hands-On Data Visualization(整理完成)"]
  n49["Reading"]
  n0 --> n4
  n3 --> n0
  n3 --> n1
  n3 --> n2
  n3 --> n4
  n3 --> n5
  n5 --> n3
  n6 --> n7
  n6 --> n8
  n6 --> n9
  n6 --> n10
  n6 --> n11
  n6 --> n12
  n6 --> n13
  n6 --> n14
  n6 --> n15
  n6 --> n16
  n6 --> n17
  n6 --> n18
  n6 --> n19
  n6 --> n20
  n21 --> n22
  n21 --> n23
  n21 --> n24
  n21 --> n25
  n21 --> n26
  n21 --> n27
  n21 --> n28
  n21 --> n29
  n21 --> n30
  n22 --> n21
  n30 --> n22
  n48 --> n31
  n48 --> n32
  n48 --> n33
  n48 --> n34
  n48 --> n35
  n48 --> n36
  n48 --> n37
  n48 --> n38
  n48 --> n39
  n48 --> n40
  n48 --> n41
  n48 --> n42
  n48 --> n43
  n48 --> n44
  n48 --> n45
  n48 --> n46
  n48 --> n47
  n49 --> n3
  n49 --> n6
  n49 --> n21
  n49 --> n48
  click n0 "../../curious-incident/ch-0001/" "ch-0001 凶案、侦探与母亲的信(第 2–163 章)"
  click n1 "../../curious-incident/ch-0002/" "ch-0002 坦白、出逃与和解(第 163–233 章)"
  click n2 "../../curious-incident/characters/" "The Curious Incident of the Dog in the Night-Time 人物"
  click n3 "../../curious-incident/" "The Curious Incident of the Dog in the Night-Time(整理完成)"
  click n4 "../../curious-incident/notes/" "The Curious Incident of the Dog in the Night-Time 笔记"
  click n5 "../../curious-incident/storyline/" "The Curious Incident of the Dog in the Night-Time 故事线"
  click n6 "../" "E-Commerce Bench(整理完成)"
  click n7 "../notes/" "E-Commerce Bench 笔记"
  click n8 "./" "part-0001 摘要与引言:一年期电商运营基准(Abstract + §1)"
  click n9 "../part-0002/" "part-0002 相关工作与定位:continuing 任务谱系(Table 1 + §2)"
  click n10 "../part-0003/" "part-0003 基准设计:四层架构与确定性经济(§3 主体)"
  click n11 "../part-0004/" "part-0004 实验结果:18 模型的多维画像(Table 2 + §4)"
  click n12 "../part-0005/" "part-0005 结论与参考文献(§5 + References)"
  click n13 "../part-0006/" "part-0006 工具集与 agent harness:18 工具与上下文管理(Table 3 + Appendix A)"
  click n14 "../part-0007/" "part-0007 数据层:类目、店型、日历与供应商(Appendix B)"
  click n15 "../part-0008/" "part-0008 经济引擎:13 步结算与需求/退货/成本公式(Table 7 + Appendix C)"
  click n16 "../part-0009/" "part-0009 确定性谈判内核:决策函数与骗局目录(Appendix D)"
  click n17 "../part-0010/" "part-0010 评测指标定义:六个维度与统计口径(Appendix E)"
  click n18 "../part-0011/" "part-0011 每维度结果导览与模型/实验设置(Table 10 + Appendix F)"
  click n19 "../part-0012/" "part-0012 失败案例研究:骗局实录与一次破产(Appendix G)"
  click n20 "../part-0013/" "part-0013 失败模式规则表与提示词设计(Table 18 + Appendix H)"
  click n21 "../../end-of-software-engineering/" "The End of Software Engineering(整理完成)"
  click n22 "../../end-of-software-engineering/notes/" "The End of Software Engineering 笔记"
  click n23 "../../end-of-software-engineering/part-0001/" "part-0001 摘要与引言:范式重构的宣告(Abstract + §1 前段)"
  click n24 "../../end-of-software-engineering/part-0002/" "part-0002 第一性原理:传统软件与 Agent 系统的形式化模型(§1 后段 + §2)"
  click n25 "../../end-of-software-engineering/part-0003/" "part-0003 三代交付史与 AI→Software→Result 的失败(§3.1–3.2)"
  click n26 "../../end-of-software-engineering/part-0004/" "part-0004 Agent→Result 与 Agentic Engineering 学科(§3.3 + §4)"
  click n27 "../../end-of-software-engineering/part-0005/" "part-0005 实证证据与 EvoClaw 落差(§5)"
  click n28 "../../end-of-software-engineering/part-0006/" "part-0006 四阶段演进路线图(前段:表 3 + Stage I–III)"
  click n29 "../../end-of-software-engineering/part-0007/" "part-0007 路线图后段与建议:实践者与研究者(§6.3 后段 + §6.4 + §7.1–7.2)"
  click n30 "../../end-of-software-engineering/part-0008/" "part-0008 组织建议、结论与参考文献(§7.3 + §8)"
  click n31 "../../hands-on-data-visualization/ch-0001/" "ch-0001 Introduction(为什么做数据可视化)"
  click n32 "../../hands-on-data-visualization/ch-0002/" "ch-0002 选工具讲你的数据故事(Ch 1)"
  click n33 "../../hands-on-data-visualization/ch-0003/" "ch-0003 强化电子表格技能(Ch 2)"
  click n34 "../../hands-on-data-visualization/ch-0004/" "ch-0004 找到并质询你的数据(Ch 3)"
  click n35 "../../hands-on-data-visualization/ch-0005/" "ch-0005 清洗脏数据(Ch 4)"
  click n36 "../../hands-on-data-visualization/ch-0006/" "ch-0006 做有意义的比较(Ch 5)"
  click n37 "../../hands-on-data-visualization/ch-0007/" "ch-0007 图表化你的数据(Ch 6)"
  click n38 "../../hands-on-data-visualization/ch-0008/" "ch-0008 地图化你的数据(Ch 7)"
  click n39 "../../hands-on-data-visualization/ch-0009/" "ch-0009 表格化你的数据(Ch 8)"
  click n40 "../../hands-on-data-visualization/ch-0010/" "ch-0010 嵌入网页(Ch 9)"
  click n41 "../../hands-on-data-visualization/ch-0011/" "ch-0011 用 GitHub 编辑与托管代码(Ch 10)"
  click n42 "../../hands-on-data-visualization/ch-0012/" "ch-0012 Chart.js 与 Highcharts 模板(Ch 11)"
  click n43 "../../hands-on-data-visualization/ch-0013/" "ch-0013 Leaflet 地图模板(Ch 12)"
  click n44 "../../hands-on-data-visualization/ch-0014/" "ch-0014 转换你的地图数据(Ch 13)"
  click n45 "../../hands-on-data-visualization/ch-0015/" "ch-0015 识别谎言、减少偏差(Ch 14)"
  click n46 "../../hands-on-data-visualization/ch-0016/" "ch-0016 讲述并展示你的数据故事(Ch 15)"
  click n47 "../../hands-on-data-visualization/ch-0017/" "ch-0017 附录 A 排查常见问题"
  click n48 "../../hands-on-data-visualization/" "Hands-On Data Visualization(整理完成)"
  click n49 "../../" "Reading"