Update README.md

README.md

@@ -1,42 +1,53 @@
 ---
 license: agpl-3.0
 ---
-# CommandNet Military Science Dataset (1900-1999)
+# COMMANDNET Military Science Dataset (1900-1999)
 
-This repository contains a synthetic, instruction-tuning dataset generator and a generated dataset built for **Unsloth fine-tuning** workflows.
+<p align="left">
+  <img src="https://img.shields.io/badge/Format-ShareGPT_JSONL-111827?style=for-the-badge&logo=json&logoColor=white" alt="Format: ShareGPT JSONL" />
+  <img src="https://img.shields.io/badge/Language-English-1f2937?style=for-the-badge" alt="Language: English" />
+  <img src="https://img.shields.io/badge/Time%20Span-1900--1999-0f766e?style=for-the-badge" alt="Time span: 1900-1999" />
+  <img src="https://img.shields.io/badge/Rows-10k-7c3aed?style=for-the-badge" alt="Rows: 10k" />
+</p>
 
-The dataset is designed around a strict persona voice (ruthless, militaristic, tactical, arrogant, analytical) and focuses on **historical/doctrinal military analysis** with explicit:
+This dataset contains synthetic instruction-tuning examples for historical and doctrinal military analysis.
+
+## Dataset Summary
+
+- Task: chat instruction tuning
+- Domain: historical military analysis
+- Time span: 1900-1999
+- Format: ShareGPT JSONL
+- Languages: English
+- Size: 10,000 rows in the included generated artifact
+- License: See repository or dataset hosting metadata
+
+It focuses on a consistent strategic voice and explicit:
 
 - Causal Analysis
 - Counterfactual Analysis
 
-It includes both conventional warfare and asymmetric contexts (insurgencies, militias, and counterinsurgency vs larger conventional forces), constrained to the historical window **1900-1999**.
+It covers both conventional and asymmetric contexts, constrained to the historical window 1900-1999.
 
----
+## Splits
 
-## Dataset Goals
-
-- Produce **1,000+ samples** (default: 1,400)
-- Maintain a consistent strategic persona style
-- Cover core military science dimensions in depth:
-  - operational art
-  - logistics and sustainment
-  - command and control (C2)
-  - intelligence cycle and ISR pressure
-  - deception and disruption
-  - maneuver vs attrition dynamics
-  - civil-military and legitimacy constraints
-  - tempo and decision-cycle pressure
-- Include **asymmetric warfare** in substantial volume
-- Keep all samples anchored to historical doctrine periods between 1900 and 1999
+- train: single full JSONL file
+- validation: not provided
+- test: not provided
 
----
+## Format
+
+Each record is a JSON object with a ShareGPT-style conversation and metadata including:
 
-## Output Schema (Default: ShareGPT)
+- year
+- decade
+- warfare_type
+- doctrine_family
+- military_science_tags
 
-The default output (`--schema sharegpt`) is directly suitable for common Unsloth chat fine-tuning pipelines.
+## Record Structure
 
-Each line is a JSON object like:
+Example:
 
 ```json
 {
@@ -56,35 +67,9 @@ Each line is a JSON object like:
 }
 ```
 
-## Record Design (Raw Schema)
-
-- `id`
-- `year`
-- `decade`
-- `year_range`
-- `warfare_type` (`conventional`, `insurgency`, `counterinsurgency`, `hybrid`)
-- `doctrine_family`
-- `force_asymmetry_index`
-- `scenario_context`
-  - `terrain`
-  - `weather`
-  - `objective`
-  - `adversary_posture`
-  - `constraint`
-- `prompt`
-- `response`
-- `causal_analysis`
-- `counterfactual_analysis`
-- `style_tags`
-- `military_science_tags`
-- `source_refs`
-- `quality_flags`
+## Included Material
 
----
-
-## Doctrine Coverage
-
-The generator samples from doctrine families including:
+The dataset is organized around doctrine families such as:
 
 - Industrial Attrition and Trench Penetration
 - Infiltration and Decentralized Assault Groups
@@ -93,59 +78,18 @@ The generator samples from doctrine families including:
 - Amphibious Operational Sequencing
 - Protracted People's War
 - Population-Centric Counterinsurgency
-- Maneuver Warfare and Decision-Cycle Pressure (OODA-linked)
+- Maneuver Warfare and Decision-Cycle Pressure
 - AirLand Battle
 - Deterrence and Escalation Management
 
-Each doctrine has active year bounds and compatible warfare types.
-
----
-
-## Quality Controls
-
-Built-in checks include:
+## Quality Characteristics
 
-- Section presence checks in responses:
-  - `Causal Analysis:`
-  - `Counterfactual Analysis:`
-  - `Command Verdict:`
-- Historical bound checks (`1900-1999`)
-- Minimum response-length check
-- Duplicate fingerprint suppression using scenario keys:
-  - year
-  - warfare type
-  - doctrine
-  - terrain
-  - objective
-  - adversary posture
+- Historical bound checks within 1900-1999
+- Explicit causal and counterfactual sections in the assistant text
+- Duplicate suppression using scenario fingerprints
+- ShareGPT-style formatting for chat fine-tuning workflows
 
-If duplicates become excessive, generation fails fast with a clear error.
+## Citation
 
----
-
-## Unsloth Integration Notes
-
-For chat SFT with Unsloth, default `sharegpt` output is the intended path.
-
-Typical flow:
+If you use this dataset, cite the dataset card and any downstream work built from it according to your project requirements.
 
-1. Load JSONL in your training notebook/script.
-2. Map `conversations` according to your model chat template.
-3. Keep the system prompt if you want style-locking behavior.
-4. Train with your normal Unsloth SFT trainer config.
-
-If your pipeline expects `instruction/input/output`, use `--schema alpaca`.
-
----
-
-## Safety and Scope
-
-This dataset is for **historical and doctrinal analysis simulation**. It is not intended to provide real-world modern operational attack guidance.
-
-Design constraints intentionally emphasize:
-
-- historical framing
-- abstract operational reasoning
-- doctrinal critique and counterfactuals
-
----