Update README.md

README.md

@@ -16,7 +16,7 @@ tags:
   - raas
 pretty_name: CYB005 — Synthetic Ransomware Attack Simulation (Sample)
 size_categories:
-  - 1K<n<10K
+  - 10K<n<100K
 ---
 
 # CYB005 — Synthetic Ransomware Attack Simulation Dataset (Sample)
@@ -28,6 +28,13 @@ Simulation Dataset** product. It contains roughly **~10% of the full
 dataset** at identical schema, actor-tier distribution, and statistical
 fingerprint, so you can evaluate fit before licensing the full product.
 
+> 🤖 **Trained baseline available:**
+> [**xpertsystems/cyb005-baseline-classifier**](https://huggingface.co/xpertsystems/cyb005-baseline-classifier)
+> — XGBoost + PyTorch MLP for **4-tier threat-actor attribution** (the
+> README's stated headline use case), group-aware split by campaign,
+> multi-seed evaluation (ROC-AUC 0.853 ± 0.031), honest leakage audit
+> of every per-timestep feature.
+
 *Note: This sample is intentionally larger than the other CYB SKU samples.
 CYB005 benchmarks are conditional on small actor-tier subsets (e.g.
 nation_state campaigns are ~10% of the fleet), so a larger sample is needed
@@ -61,6 +68,27 @@ state machine** across enterprise, cloud, and OT/ICS environments, with:
 - **Living-off-the-Land (LotL)** abuse and EDR signature lag modeling
 - **Financial impact scoring** with ransom demand × payment probability
 
+## Trained Baseline Available
+
+A working baseline classifier trained on this sample is published at
+**[xpertsystems/cyb005-baseline-classifier](https://huggingface.co/xpertsystems/cyb005-baseline-classifier)**.
+
+| Component | Detail |
+|---|---|
+| Task | **4-class threat-actor capability-tier attribution** (the README's headline use case) |
+| Models | XGBoost (`model_xgb.json`) + PyTorch MLP (`model_mlp.safetensors`) |
+| Features | 63 (after one-hot encoding); pipeline included as `feature_engineering.py` |
+| Split | **Group-aware by campaign_id** — train/val/test campaigns disjoint |
+| Validation | Single seed + multi-seed aggregate across 10 seeds |
+| Demo | `inference_example.ipynb` — end-to-end copy-paste |
+| Headline metrics | XGBoost: accuracy 0.603 ± 0.040, macro ROC-AUC 0.853 ± 0.031 (multi-seed) |
+
+This is the **first XpertSystems baseline to ship the dataset's stated
+headline use case** (rather than pivoting to a phase-prediction subtask
+as the smaller CYB002 / CYB003 / CYB004 samples required). CYB005's
+500-campaign sample is large enough that tier attribution learns
+honestly under group-aware splitting.
+
 ## Calibrated Benchmark Targets
 
 The full product is calibrated to 12 benchmark metrics drawn from
@@ -137,7 +165,8 @@ log and segment registry schemas respectively.
 
 ## Suggested Use Cases
 
-- Training **ransomware classifier** models (4-tier actor attribution)
+- Training **ransomware classifier** models —
+  [worked example available](https://huggingface.co/xpertsystems/cyb005-baseline-classifier)
 - **Backup posture risk modeling** — predict recovery likelihood from
   6-tier backup maturity
 - **Dwell time forecasting** under varying actor capability and defender
@@ -175,6 +204,10 @@ y_recovered = summaries["recovery_without_payment_flag"]
 y_wiper  = summaries["wiper_component_flag"]
 ```
 
+For a worked end-to-end example with actor-tier classification,
+group-aware splitting, and feature engineering, see the inference notebook
+in the [baseline classifier repo](https://huggingface.co/xpertsystems/cyb005-baseline-classifier/blob/main/inference_example.ipynb).
+
 ## License
 
 This **sample** is released under **CC-BY-NC-4.0** (free for non-commercial