Update README.md
Browse files
README.md
CHANGED
|
@@ -1,10 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
-
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
---
|
| 9 |
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# SORTMARK Official Laboratory (空间计算与边缘安全小模型研究)
|
| 2 |
+
|
| 3 |
+
<p align="center">
|
| 4 |
+
<a href="https://sortmark.com">🌐 Official Website 官方网站</a> |
|
| 5 |
+
<a href="mailto:contact@sortmark.com">📬 Contact 业务联络</a> |
|
| 6 |
+
<a href="https://github.com/sortmark-lab">🐙 GitHub Organization</a>
|
| 7 |
+
</p>
|
| 8 |
+
|
| 9 |
+
<p align="center">
|
| 10 |
+
<img src="https://img.shields.io/badge/Focus-Edge%20SLM%20%26%20Spatial%20Computing-cyan.svg" alt="Focus" />
|
| 11 |
+
<img src="https://img.shields.io/badge/Format-GGUF%20Q4__K__M-success.svg" alt="GGUF" />
|
| 12 |
+
<img src="https://img.shields.io/badge/Inference-Pure%20Offline%20CPU-brightgreen.svg" alt="Offline Inference" />
|
| 13 |
+
<img src="https://img.shields.io/badge/License-Apache--2.0-blue.svg" alt="License" />
|
| 14 |
+
</p>
|
| 15 |
+
|
| 16 |
---
|
| 17 |
+
|
| 18 |
+
## 🌐 English Overview
|
| 19 |
+
|
| 20 |
+
**SORTMARK** is dedicated to high-precision spatial computing, robotic kinematics, and specialized Small Language Models (SLMs) for edge environments. We design lightweight, deterministic models that compile business intents into schema-compliant security policies without cloud dependencies.
|
| 21 |
+
|
| 22 |
+
### 🎯 Key Focus Areas
|
| 23 |
+
- **Edge Policy Compilation**: Translating global trade and network traffic intents into deterministic Zero-Trust Policy-as-Code (PaC) rules.
|
| 24 |
+
- **Edge Telemetry Data Audit**: Delivering structured compliance inspection, PII identification, and dynamic masking for distributed supply chain nodes.
|
| 25 |
+
- **Pure Offline Inference**: Deployable on low-power edge CPUs (<500MB RAM footprint) using Ollama and llama.cpp.
|
| 26 |
+
- **Stakeholder Synergy**: Aiding CIOs, CTOs, DPOs, Security Architects, Information Engineers, and Business Analysts (BAs).
|
| 27 |
+
|
| 28 |
+
---
|
| 29 |
+
|
| 30 |
+
## 🇨🇳 中文简介
|
| 31 |
+
|
| 32 |
+
**SORTMARK** 专注于高精度空间计算、机器人多轴机构原理与端侧轻量小模型(SLM)研发。我们致力于打造兼具确定性规则约束与极简离线推理能力的边缘智能资产,无需依赖公有云即可在工业控制台与本地网络安全运行。
|
| 33 |
+
|
| 34 |
+
### 🌟 核心研究与交付矩阵
|
| 35 |
+
- **网络安全策略模型**:将跨国贸易与供应链访问意图,转化为严格符合 JSON Schema 规范的防火墙与 ACL 规则。
|
| 36 |
+
- **边缘遥测合规审计**:对全球出海业务中的日志报文、PII 个人隐私标识与字段脱敏提供结构化辅助审查。
|
| 37 |
+
- **纯本地 CPU 推理**:模型体积约 379 MB(Q4_K_M GGUF),内存占用低于 500 MB,无需独立显卡即可稳定运行。
|
| 38 |
+
- **赋能决策体系**:为组织高管决策、CIO、CTO、DPO(数据保护官)、安全架构师、信息工程团队与业务分析师(BA)提供专业洞察。
|
| 39 |
+
|
| 40 |
+
---
|
| 41 |
+
|
| 42 |
+
## 🇯🇵 日本語概要
|
| 43 |
+
|
| 44 |
+
**SORTMARK** は、高精度空間コンピューティング、ロボティクス運動学、およびエッジ環境特化型スモール言語モデル(SLM)の研究開発を行っています。クラウド依存を排除し、完全なオフライン環境で確定的かつ構造化されたポリシーを生成する技術を提供します。
|
| 45 |
+
|
| 46 |
+
### 🚀 主な取り組み
|
| 47 |
+
- **エッジ向けセキュリティポリシー生成**: グローバル取引のアクセス意図を、確定的なゼロトラストPaC規則に変換。
|
| 48 |
+
- **テレメトリデータ監査**: 国際物流やエッジノードにおけるPII識別、データコンプライアンス検証、動的マスキング案の提示。
|
| 49 |
+
- **CPU単体オフライン実行**: GGUF形式により、一般的なエッジ端末(RAM 500MB未満)で安全・高速に稼働。
|
| 50 |
+
- **対象ステークホルダー**: CIO、CTO、DPO、セキュリティアーキテクト、情報工学エンジニア、BA。
|
| 51 |
+
|
| 52 |
---
|
| 53 |
|
| 54 |
+
## 🔬 Evaluation Notice & Community Feedback (评估版本说明与反馈)
|
| 55 |
+
> ⚠️ **Evaluation Notice (非生产环境建议)**:
|
| 56 |
+
> All models published by SORTMARK are experimental evaluation releases (**v0.1-beta**). Provided **"AS IS"** for academic exploration, architectural benchmarking, and feasibility testing. **Not recommended for direct production deployment.** We warmly welcome community feedback, issue reports, and suggestions!
|
| 57 |
+
> 📬 *Contact & Feedback*: [contact@sortmark.com](mailto:contact@sortmark.com) | [https://sortmark.com](https://sortmark.com)
|