RinaldiDev/flan_paradetox_full

FLAN-T5-Large Polite-Rewrite (full fine-tune)

Model: google/flan-t5-large fine-tuned for toxic → polite rewriting.

Training details

Parameter	Value
epochs	3
effective batch	32 (16 × grad_acc=2, fp16)
lr / schedule	3 e-5, cosine, 3 % warm-up
total steps	1 800
optimizer	AdamW, weight_decay=0.01
hardware	1 × A100-40 GB

Data

Merged 29 k parallel pairs

• ParaDetox (19 k)
• Polite Insult (1.6 k, oversample×2)
• PseudoParaDetox Llama-3 (8.6 k, tox≤0.3, cosine≥0.8)

Metrics (dev 3 %)

metric	score
BLEU	0.82
Avg toxicity (Detoxify)	0.12 (src 0.71 → tgt 0.12)
Success rate (tox≤0.5 AND -20 %)	89 %

Usage

from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tok = AutoTokenizer.from_pretrained("RinaldiDev/flan-paradetox-full")
model = AutoModelForSeq2SeqLM.from_pretrained("RinaldiDev/flan-paradetox-full")

def rewrite_polite(text):
    inp = f"Rewrite politely:\\nInput: {text}\\nPolite:"
    ids = tok(inp, return_tensors="pt").input_ids
    out = model.generate(ids, num_beams=4, max_length=96)
    return tok.decode(out[0], skip_special_tokens=True)

print(rewrite_polite("Shut up, idiot!"))
# → "Stop talking"

<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->

### Direct Use

<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->


AI moderation helper
Toxic-to-polite assistants
Not for hallucination-free tasks; may still miss subtle hate speech.

### Downstream Use [optional]

<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->

[More Information Needed]

### Out-of-Scope Use

<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->

[More Information Needed]

## Bias, Risks, and Limitations

<!-- This section is meant to convey both technical and sociotechnical limitations. -->


Trained largely on English; fails on code-switching.
Llama-generated pairs could contain artifacts.

### Recommendations

<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.

## How to Get Started with the Model

Use the code below to get started with the model.

[More Information Needed]

## Training Details

### Training Data

<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->

[More Information Needed]

### Training Procedure

<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->

#### Preprocessing [optional]

[More Information Needed]


#### Training Hyperparameters

- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->

#### Speeds, Sizes, Times [optional]

<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->

[More Information Needed]

## Evaluation

<!-- This section describes the evaluation protocols and provides the results. -->

### Testing Data, Factors & Metrics

#### Testing Data

<!-- This should link to a Dataset Card if possible. -->

[More Information Needed]

#### Factors

<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->

[More Information Needed]

#### Metrics

<!-- These are the evaluation metrics being used, ideally with a description of why. -->

[More Information Needed]

### Results

[More Information Needed]

#### Summary



## Model Examination [optional]

<!-- Relevant interpretability work for the model goes here -->

[More Information Needed]

## Environmental Impact

<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->

Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).

- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]

## Technical Specifications [optional]

### Model Architecture and Objective

[More Information Needed]

### Compute Infrastructure

[More Information Needed]

#### Hardware

[More Information Needed]

#### Software

[More Information Needed]

## Citation [optional]

<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->

**BibTeX:**

[More Information Needed]

**APA:**

[More Information Needed]

## Glossary [optional]

<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->

[More Information Needed]

## More Information [optional]

[More Information Needed]

## Model Card Authors [optional]

[More Information Needed]

## Model Card Contact

[More Information Needed]