Arbiter — T5 cybersecurity Q&A
What this is
A T5 conditional-generation checkpoint fine-tuned for Arbiter.
Used for general cybersecurity questions (phishing, ransomware, DDoS, etc.) when the input is not classified as a structured log. The app returns a short generated answer via T5ForConditionalGeneration.generate(...).
Project
Part of Arbiter (Django + React): paste a security log or ask a cybersecurity question.
This repository holds weights only. Application code: github.com/SentinelSage/arbiter.
Training note
This is a custom fine-tune, not an unmodified public T5 base/small checkpoint. A generic Hub T5 model is not a guaranteed substitute for this artifact.
Training data: custom-curated, unpublished project data. Training tables are not published with the app.
Intended use
- Running the Arbiter full-ML path locally
- Experimentation and educational use with the companion application
Not a production SOC product. No warranty. Generated text can be incomplete or incorrect; do not use for live security operations without your own validation.
How Arbiter loads it
Local path after download:
backend/models/cyber_qa_t5_model_general/
Loaded with T5ForConditionalGeneration.from_pretrained(<local_or_hub_id>) when ARBITER_DEMO_MODE=false.
Files
Expect a standard Transformers export, e.g. model.safetensors, config.json, spiece.model, tokenizer config files.
Method / citations
- T5: Raffel, C., et al. (2020). Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer. JMLR.
T5 docs - Transformers: Wolf, T., et al. (2020). Transformers: State-of-the-Art Natural Language Processing. EMNLP.
huggingface/transformers - PyTorch: Paszke, A., et al. (2019). PyTorch: An Imperative Style, High-Performance Deep Learning Library. NeurIPS.
pytorch/pytorch
Safety
Research / portfolio demo only. Do not paste production logs, PII, or credentials into shared demos using these weights.
License
Apache License 2.0. Demo only — provided as-is, without warranty of any kind.
Copyright 2026 Ryan Vong / Sentinel Sage