From d023e3743efe2b16fe6249bcaf88a2fd016ad942 Mon Sep 17 00:00:00 2001 From: Junhao Hu Date: Fri, 19 Jun 2026 10:36:13 +0800 Subject: [PATCH] Expand CRM/ERP acronyms in README --- README.md | 2 +- README.zh.md | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index fa2dd4f9..76c14e78 100644 --- a/README.md +++ b/README.md @@ -2,7 +2,7 @@ Languages: [English](README.md) | [Chinese](README.zh.md) -**GDPevo** is a public benchmark for evaluating agent self-evolution on real business work. The release contains 120 tasks across 12 task groups in CRM, ERP, and Finance; each group has one shared business environment, 5 train tasks, and 5 held-out test tasks. For the full motivation, construction pipeline, and findings, read the [project blog](https://prism-shadow.github.io/GDPevo/blog.html). +**GDPevo** is a public benchmark for evaluating agent self-evolution on real business work. The release contains 120 tasks across 12 task groups in customer relationship management (CRM), enterprise resource planning (ERP), and Finance; each group has one shared business environment, 5 train tasks, and 5 held-out test tasks. For the full motivation, construction pipeline, and findings, read the [project blog](https://prism-shadow.github.io/GDPevo/blog.html). ## Evaluation Results diff --git a/README.zh.md b/README.zh.md index 75b39414..00317f1c 100644 --- a/README.zh.md +++ b/README.zh.md @@ -2,7 +2,7 @@ 语言：[English](README.md) | [中文](README.zh.md) -**GDPevo** 是一个公开基准，用来评估智能体在真实企业任务上的自进化能力。当前版本包含 120 个任务，覆盖客户关系管理、企业资源计划和金融三类业务，共 12 个任务组；每个任务组都有一个共享业务环境、5 个训练任务和 5 个保留测试任务。完整动机、构建流程和结果分析见[项目博客](https://prism-shadow.github.io/GDPevo/blog.html)。 +**GDPevo** 是一个公开基准，用来评估智能体在真实企业任务上的自进化能力。当前版本包含 120 个任务，覆盖客户关系管理（CRM）、企业资源计划（ERP）和金融三类业务，共 12 个任务组；每个任务组都有一个共享业务环境、5 个训练任务和 5 个保留测试任务。完整动机、构建流程和结果分析见[项目博客](https://prism-shadow.github.io/GDPevo/blog.html)。 ## 评测结果 @@ -26,7 +26,7 @@ | 路径 | 内容 | | --- | --- | -| [`data/`](data/) | 已发布的基准数据，包括任务组、共享环境、训练与测试任务、参考答案和规则评测器。 | +| [`data/`](data/) | 已发布的基准数据，包括任务组、共享环境、训练与测试任务、参考答案和基于规则的评测器。 | | [`experiments/`](experiments/) | 已发布的评测结果、可复用评测工作区、报告文件和实验汇总表。 | | [`site/`](site/) | 基准发布用的公开网站与博客。 |