Update README.md

README.md

@@ -1,38 +1,74 @@
 ---
 license: mit
 task_categories:
-- text-generation
+  - text-generation
 tags:
-- code
-- agents
-- program-reproduction
+  - code
+  - agents
+  - program-reproduction
 ---
 
-# programgym-1648
+# 🧩 ProgramGym-1648
 
-[Paper](https://huggingface.co/papers/2610.13395) | [Project page](https://programgym.github.io/) | [Code](https://github.com/Acm1011/ProgramGym)
+**A dataset of 1,648 tasks for long-horizon program reproduction agents.**
 
-programgym-1648 is a collection of 1,648 Harbor tasks that evaluate an agent's ability to reproduce software from observed behavior. For each task, the agent receives a reference executable and documentation. It must create an original codebase and a `compile.sh` script that builds a new `./executable` with matching behavior. Copying, wrapping, or decompiling the reference executable is not allowed.
+[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](https://opensource.org/licenses/MIT)
+![Tasks](https://img.shields.io/badge/Tasks-1%2C648-brightgreen)
+![Harbor](https://img.shields.io/badge/Harbor-%E2%89%A5%200.20.0-orange)
+![Task Schema](https://img.shields.io/badge/Task_Schema-1.3-purple)
 
-## Task structure
+📄 **[Paper](https://huggingface.co/papers/2610.13395)** · 🌐 **[Project Page](https://programgym.github.io/)** · 💻 **[GitHub](https://github.com/Acm1011/ProgramGym)**
 
-Each task contains:
+---
+
+## 📖 Overview
+
+**ProgramGym-1648** is a collection of **1,648 Harbor tasks** designed to train an AI agent's ability to reproduce software programs from their observed behavior.
+
+In each task, the agent is given a **reference executable** and accompanying **documentation**. The objective is to independently reconstruct the program's functionality by developing a new codebase that matches the reference program's behavior.
+
+The agent must deliver:
+
+- 📝 **Source Code** — An original implementation of the target program.
+- 🔧 **Build Script** — A `compile.sh` script that compiles the implementation.
+
+> 🚫 **Important:** Copying, wrapping, or decompiling the reference executable is strictly prohibited. Agents must reproduce the program's functionality through their own implementations.
+
+## 📂 Task Structure
+
+Each task follows a standardized Harbor-compatible directory structure:
+
+| Path | Description |
+|---|---|
+| 📄 `task.toml` | Harbor task metadata and runtime configuration |
+| 📋 `instruction.md` | Task requirements, instructions, and constraints |
+| 🖥️ `environment/` | Agent execution environment and reference executable |
+| 🧪 `tests/` | Verification logic and test fixtures |
+| ✅ `solution/` | Harbor solution entry point for validation |
+
+## 🚀 Quick Start
+
+ProgramGym-1648 can be evaluated directly using [Harbor](https://github.com/harbor-framework/harbor).
+
+### 1. Prerequisites
 
-| Path             | Purpose                                         |
-| ---------------- | ----------------------------------------------- |
-| `task.toml`      | Harbor task metadata and runtime settings.      |
-| `instruction.md` | The task requirements and rules.                |
-| `environment/`   | The agent environment and reference executable. |
-| `tests/`         | The verifier and test fixtures.                 |
-| `solution/`      | Harbor's solution entry point for validation.   |
+Before running the benchmark, make sure you have:
 
-## Run with Harbor
+- 🐳 **Docker** installed and running.
+- 🔑 A configured model provider with valid API credentials.
+- 🛠️ **Harbor ≥ 0.20.0**, supporting task schema 1.3 and phase-scoped network policies.
 
-Requires Docker, a configured model provider, and **Harbor 0.20.0 or newer**. These tasks use Harbor task schema 1.3 and phase-scoped network policies. Run it by Harbor:
+Install Harbor using `uv`:
 
 ```bash
 uv tool install 'harbor>=0.20.0'
+```
+
+### 2. Run the Benchmark
+
+Launch the full benchmark with the following command:
 
+```bash
 harbor run \
   -d ProgramGym/programgym-1648 \
   -a claude-code \
@@ -40,20 +76,37 @@ harbor run \
   -n 4
 ```
 
-Replace the model ID and provide the corresponding API credentials. The command runs the full dataset with four concurrent trials; use `-l 1` for a one-task smoke test. Results are written under `jobs/` by default.
+**Configuration:**
 
-## Citation
+| Argument | Description |
+|---|---|
+| `-d` | Dataset identifier |
+| `-a` | Agent implementation (e.g., `claude-code`) |
+| `-m` | Model provider and model identifier |
+| `-n` | Number of concurrent trials (`4` in this example) |
 
-If you find our work helpful, feel free to give us a cite.
+Replace `your-provider/your-model` with your target model and configure the corresponding API credentials.
 
-```
+### 3. Results
+
+📊 By default, Harbor writes execution logs and results to the `jobs/` directory.
+
+## 📚 Citation
+
+If you find **ProgramGym** useful in your research, please consider citing our paper:
+
+```bibtex
 @misc{yang2026programgym,
-      title={ProgramGym: Scaling Training Environments for Long-Horizon Program Reproduction Agents},
-      author={Chengyi Yang and Renxu Bao and Jue Hou and Zongpei Teng and Ziheng Wang and Yujie Lin and Chengsong Huang and Junnan Zhu and Delai Qiu and Yining Wang and Shengping Liu and Fei Long and Jinsong Su},
-      year={2026},
-      eprint={2610.13395},
-      archivePrefix={arXiv},
-      primaryClass={cs.SE},
-      url={https://arxiv.org/abs/2610.13395},
+  title         = {ProgramGym: Scaling Training Environments for Long-Horizon Program Reproduction Agents},
+  author        = {Chengyi Yang and Renxu Bao and Jue Hou and Zongpei Teng and Ziheng Wang and Yujie Lin and Chengsong Huang and Junnan Zhu and Delai Qiu and Yining Wang and Shengping Liu and Fei Long and Jinsong Su},
+  year          = {2026},
+  eprint        = {2610.13395},
+  archivePrefix = {arXiv},
+  primaryClass  = {cs.SE},
+  url           = {https://arxiv.org/abs/2610.13395}
 }
-```
+```
+
+## 📜 License
+
+This dataset is released under the **[MIT License](https://opensource.org/licenses/MIT)**.