Add dataset card for EC²-Bench

README.md

@@ -0,0 +1,23 @@
+---
+license: mit
+task_categories:
+- text-generation
+tags:
+- agent
+- edge-cloud
+- benchmark
+---
+
+# EC²-Bench
+
+**Empowering Edge Agents: A Systematic Analysis of Edge-Cloud Collaboration in LLM Agent Execution**
+
+This repository contains **EC²-Bench** (Edge–Cloud Collaborative Agent Benchmark), a benchmark for evaluating **edge–cloud collaboration in long-horizon LLM agent execution**. It provides **128 executable digital-workspace tasks** across six categories—Office & Daily Tasks, Information Search & Gathering, Safety & Security, Development & Operations, Data Analysis, and Automation—with **100 tasks carrying fine-grained privacy annotations**.
+
+The benchmark measures each run on three dimensions: **task performance**, **resource usage**, and **privacy exposure**.
+
+- **Paper:** [Empowering Edge Agents: A Systematic Analysis of Edge-Cloud Collaboration in LLM Agent Execution](https://huggingface.co/papers/2610.17816)
+- **Code:** [https://github.com/OpenBMB/EC2-Bench](https://github.com/OpenBMB/EC2-Bench)
+- **Project page:** [https://passionate11.github.io/ec2-bench-project-page/](https://passionate11.github.io/ec2-bench-project-page/)
+
+For full setup instructions, task descriptions, evaluation commands, and leaderboard, please refer to the [GitHub repository](https://github.com/OpenBMB/EC2-Bench).