---
license: cc-by-nc-4.0
language: pt
tags:
  - ai-detection
  - sentence-level
  - portuguese
  - bert-base
  - transformers
pipeline_tag: text-classification
base_model: Detecting-ai/pt-ai-detector          # doc-level parent
datasets:
  - wiki40b-pt
  - oscar-pt
  - cc100-pt
  - europarl-pt
  - opus-books-pt
  - Detecting-ai/ai_pt_corpus                    # 1 M AI-written lines
model_type: bert
---

# 🇧🇷 **pt-ai-detector-sent**

Sentence-level Portuguese classifier that flags whether a *single sentence* was likely written by a Large-Language-Model (LLM) or by a human.

**Why?** The document-level model [`Detecting-ai/pt-ai-detector`](https://huggingface.co/Detecting-ai/pt-ai-detector) works great on paragraphs, but very short inputs lost accuracy.  
This checkpoint inherits that backbone and is fine-tuned on 200 k balanced sentences.

| property | value |
|----------|-------|
| **Base checkpoint** | `Detecting-ai/pt-ai-detector` |
| **Fine-tune data**  | 100 000 human + 100 000 AI sentences (≥ 4 words) |
| **LLMs used for AI text** | Azure OpenAI *gpt-4o-mini*, *gpt-4o*, *gpt-35-turbo* |
| **Training** | 1 epoch · batch 16 · lr 1 e-5 · A100-40 GB |
| **Validation F1** | **0.989** (balanced sentences) |
| **Intended use** | quick checks inside larger pipelines, sentence-by-sentence highlighting |

> **Demo** : see the free web checker at **[detecting-ai.com](https://detecting-ai.com/pt)** — powered by this model.

---

## ✨ Quick start

```python
from transformers import pipeline

clf = pipeline(
    "text-classification",
    model="Detecting-ai/pt-ai-detector-sent",
    tokenizer="Detecting-ai/pt-ai-detector-sent",
    device_map="auto"          # GPU if available
)

txt = "A inteligência artificial está transformando a educação."
print(clf(txt, top_k=None))
# → [{'label': 'LABEL_1', 'score': 0.87}]   # 1 = AI, 0 = Human
```

## 🔧 Recommended threshold

| score range | interpretation |
|-------------|----------------|
| **> 0.70**  | likely **AI** (LLM-generated / paraphrased) |
| 0.30 – 0.70 | *uncertain* – review in context |
| **< 0.30**  | likely **Human** |

*For full documents, classify every sentence and aggregate (e.g. “flag as AI if ≥ 30 % of sentences score > 0.70”).*

---

## 🗂️ Training data

| corpus | purpose |
|--------|---------|
| **wiki40b-pt**, **oscar-pt**, **cc100-pt**, **europarl-pt**, **opus-books-pt** | human prose (web, books, parliament) |
| **Detecting-ai/ai_pt_corpus** | 1 M AI sentences generated with Azure OpenAI models (news, essays, chat, tweets, dialogs, code comments) |

> All human corpora were cleaned (language-ID filter, deduplication, URL removal).  
> Sentences shorter than **4 tokens** were dropped.

---

## 📈 Validation metrics

| split  | precision | recall | F1 |
|--------|-----------|--------|----|
| **Human** | 0.987 | 0.990 | 0.989 |
| **AI**    | 0.991 | 0.988 | 0.989 |
| **Macro** | 0.989 | 0.989 | 0.989 |

*Evaluated on a held-out, balanced set of 20 k sentences.*

---

## ⚠️ Limitations & caveats

* Best on Portuguese sentences **≥ 8–10 tokens**; very short fragments are mostly noise.  
* Trained on mainstream GPT family (GPT-4o, GPT-35-turbo); accuracy may drop on entirely novel models or heavy prompt-engineering.  
* Occasional false-positives on very formal human writing; false-negatives on heavy slang AI output.  
* **Not** a plagiarism detector and does **not** guarantee authorship.

---

## 📜 License

Creative Commons **CC-BY-NC 4.0** – free for research & non-commercial use.  
Commercial use requires written permission from the authors.

---

## 🤝 Team & contact

Built with ❤️ by the team behind **[detecting-ai.com](https://detecting-ai.com)**.  
Questions, issues, partnership requests → **support@detecting-ai.com**