Description

SOC Narrative is a framework for insider threat detection using small open-weight LLMs. A model receives a user/day window of events from the CERT Insider Threat Dataset R4.2 and must produce a structured response with:

• Risk label: normal, suspicious, or malicious
• Evidence: cited event IDs supporting the decision
• Reasoning: brief explanation of the investigation logic

This project explores whether small LLMs (3B–14B) can match or exceed traditional ML baselines for UEBA (User and Entity Behavior Analytics).

Metrics

Evaluation results for this checkpoint are not yet available. See the project repo for details.

Quick Usage

python
Copy code
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base = "Qwen/Qwen3-14B"
model = AutoModelForCausalLM.from_pretrained(base, torch_dtype="auto", device_map="auto")
model = PeftModel.from_pretrained(model, "Pankei/soc-narrative-sft-smoke-qwen3-14b")

tokenizer = AutoTokenizer.from_pretrained(base)
inputs = tokenizer("<your prompt>", return_tensors="pt").to(model.device)
output = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(output[0]))

Note: This is a LoRA adapter (~30–160 MB). You need the full base model (Qwen/Qwen3-14B) to load it.

Training Details

• Base model: Qwen/Qwen3-14B
• Method: SFT LoRA
• Train data: 1 step (infrastructure test)
• Checkpoint: step 1
• LoRA rank: 32, alpha: 64, target modules: q_proj, k_proj, v_proj, o_proj
• Format: Structured SOC Narrative (risk + evidence + reasoning)
• Hardware: NVIDIA H100 (80 GB)

Limitations

• Evaluated on a small balanced sample (n=50) — results may not generalize to production distributions
• 1-step smoke test for training pipeline validation. Not intended for production use.
• Dataset is based on synthetic insider threat scenarios from CERT R4.2 (2016) — real-world performance may differ

Citation

bibtex
Copy code
@misc{soc-narrative-2026,
  author = {Research project},
  title = {SOC Narrative: Small LLMs for UEBA / Insider Threat Detection},
  year = {2026},
  howpublished = {\url{https://github.com/Pancake2021/research_work_by_a_student}}
}