Update README.md

README.md

@@ -1,199 +1,164 @@
 ---
 library_name: transformers
-tags: []
+license: openrail++
+datasets:
+- s-nlp/pseudoparadetox_llama3_70b_10shot_noabl
+language:
+- en
+base_model:
+- facebook/bart-large
+pipeline_tag: text-generation
 ---
 
-# Model Card for Model ID
-
-<!-- Provide a quick summary of what the model is/does. -->
-
-
+# Model Card for s-nlp/bart-large-pseudoparadetox-llama3-70b-10shot
 
 ## Model Details
 
-### Model Description
-
-<!-- Provide a longer summary of what this model is. -->
+This model is a BART-large sequence-to-sequence model fine-tuned for **English Text Detoxification** (style transfer from toxic to neutral). It was trained on the **PseudoParaDetox** dataset, which was synthetically generated using the Llama 3 70B Instruct LLM in a 10-shot setting, leveraging Activation Patching to bypass safety alignment during the data generation phase.
 
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
+The resulting detoxification model demonstrates high fluency and content preservation scores, outperforming models trained on the original human-annotated ParaDetox dataset in manual human evaluation.
 
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
+- **Developed by:** Daniil Moskovskiy, Sergey Pletenev, and Alexander Panchenko
+- **Model type:** Encoder-Decoder (BART-large)
+- **Language(s) (NLP):** English (en)
+- **License:** OpenRAIL++
+- **Finetuned from model:** `facebook/bart-large`
 
 ### Model Sources [optional]
 
-<!-- Provide the basic links for the model. -->
-
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
+- **Repository (Code & Data):** [https://github.com/s-nlp/pseudoparadetox](https://github.com/s-nlp/pseudoparadetox)
+- **Paper:** "LLMs to Replace Crowdsourcing For Parallel Data Creation? The Case of Text Detoxification" (Moskovskiy, Pletenev, \& Panchenko, EMNLP XXXX)
 
 ## Uses
 
-<!-- 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. -->
+The model is intended for the automatic rewriting of toxic, offensive, or rude English input text into a polite or neutral tone, while maintaining the original semantic meaning and fluency.
 
-[More Information Needed]
+*   Filtering text for online forums or social media.
+*   Enabling polite response generation in conversational agents.
+*   Assisting users in editing drafted messages to be more respectful.
 
 ### 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]
+This model can be integrated into larger content moderation pipelines, specifically handling the mitigation step after toxicity detection.
 
 ### Out-of-Scope Use
 
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+The model is trained strictly for English detoxification. It should not be used for:
 
-[More Information Needed]
+*   Generating original, creative content (it is a rewriting model).
+*   Detoxification in languages other than English.
+*   Censoring legitimate political or critical commentary that does not violate toxicity guidelines (though this is a risk inherent to all detoxification systems).
 
 ## Bias, Risks, and Limitations
 
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+This model inherits typical limitations of text style transfer models:
 
-[More Information Needed]
+*   **Semantic Loss (Over-Sanitization):** In attempts to fully remove toxicity, the model may sometimes alter or dilute the core meaning of the original statement, especially when dealing with complex or subtle insults.
+*   **Inherent Bias:** The base BART model and the LLM used for data generation (Llama 3) carry pre-existing biases, which may manifest as inconsistent detoxification quality across different demographics or topics.
+*   **Data Generation Risk:** The training data was created using an Activation Patching technique to bypass the Llama 3's safety alignment. While this was necessary for generating high-quality parallel data, users should be aware that the training data distribution might reflect content that human annotators typically refuse to handle.
 
 ### 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.
+We recommend systematic testing on new domain data before deployment. Users should implement a post-processing toxicity classifier to confirm that the detoxified output is truly non-toxic, as detoxification is not guaranteed in 100% of cases.
 
 ## How to Get Started with the Model
 
-Use the code below to get started with the model.
-
-[More Information Needed]
+Use the code below to get started with the model:
 
-## Training Details
+```python
+from transformers import pipeline
 
-### Training Data
+# Note: Replace 'Model ID' with the actual Hugging Face path once available
+detoxifier = pipeline("text2text-generation", model="s-nlp/bart-large-pseudoparadetox-llama3-70b-10shot")
 
-<!-- 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. -->
+toxic_text = "You are dumb idiot!"
 
-[More Information Needed]
+result = detoxifier(toxic_text, max_length=128)
+print(result[0]['generated_text'])
+# Expected Output: You are wrong! (or similar neutral rewrite)
+```
 
-### Training Procedure
+## Training Details
 
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+### Training Data
 
-#### Preprocessing [optional]
+The model was fine-tuned on **PseudoParaDetox**, a synthetic dataset generated by the Llama 3 70B Instruct model. The source texts were derived from the toxic side of the **ParaDetox** corpus. The generation utilized a 10-shot prompt setup to guide the LLM's rewriting process. This generation was facilitated by an Activation Patching technique to prevent the LLM from refusing to generate detoxified output for highly toxic inputs.
 
-[More Information Needed]
+### Training Procedure
 
+The BART-large model was fine-tuned using the following key parameters:
 
 #### 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]
+- **Optimizer:** AdamW
+- **Learning Rate:** 0.00005
+- **Batch Size:** 32 (with 1 gradient accumulation step)
+- **Epochs:** 5
+- **Precision:** bfloat16
 
 ## Evaluation
 
-<!-- This section describes the evaluation protocols and provides the results. -->
+The final model performance was assessed using a combination of automatic metrics (on the ParaDetox test set) and side-by-side comparisons using GPT-4o, followed by manual human evaluation.
 
 ### 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]
+ParaDetox private test split (671 texts).
 
 #### Metrics
 
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-
-[More Information Needed]
+*   **Style Transfer Accuracy (STA)**: Measured by a RoBERTa-based toxicity [classifier](https://huggingface.co/s-nlp/roberta_first_toxicity_classifier). (Higher is better, reflecting successful toxicity removal.)
+*   **Semantic Similarity (SIM)**: Measured by [BLEURT score](https://huggingface.co/Elron/bleurt-large-512) between the original and detoxified text.
+*   **Fluency (FL)**: Measured by a RoBERTa-based linguistic acceptability [classifier](https://huggingface.co/cointegrated/roberta-large-cola-krishna2020) (CoLA-trained).
+*   **Joint Score (J)**: The geometric mean of STA, SIM, and FL.
 
 ### Results
 
-[More Information Needed]
-
-#### Summary
-
+The results below reflect the performance of BART fine-tuned on the Llama 3 70B A.P. 10-shot data, compared against the baseline BART trained on the original human-annotated ParaDetox data.
 
+| Metric | BART (Original ParaDetox) | BART (PseudoParaDetox 70B AP 10-shot) |
+| :--- | :--- | :--- |
+| **STA (Auto)** | 0.876 | 0.842 |
+| **SIM (Auto)** | 0.616 | 0.594 |
+| **FL (Auto)** | 0.824 | 0.866 |
+| **Joint Score (J)** | 0.444 | 0.434 |
+| **Manual J Score** | 0.661 | **0.762** |
+| **GPT-4o Win Rate vs. Baseline** | - | **65\%** |
 
-## 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
+#### Summary
 
-[More Information Needed]
+While automatic metrics show comparable performance, the superior Manual Joint Score (J=0.762) and high GPT-4o side-by-side win rate (65\%) indicate that the data generated using the patched LLM results in subjectively higher-quality detoxification compared to the original crowdsourced data.
 
 ## 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]
+```bibtex
+@inproceedings{moskovskiy-etal-2024-llms,
+    title = "{LLM}s to Replace Crowdsourcing For Parallel Data Creation? The Case of Text Detoxification",
+    author = "Moskovskiy, Daniil  and
+      Pletenev, Sergey  and
+      Panchenko, Alexander",
+    editor = "Al-Onaizan, Yaser  and
+      Bansal, Mohit  and
+      Chen, Yun-Nung",
+    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2024",
+    month = nov,
+    year = "2024",
+    address = "Miami, Florida, USA",
+    publisher = "Association for Computational Linguistics",
+    url = "https://aclanthology.org/2024.findings-emnlp.839/",
+    doi = "10.18653/v1/2024.findings-emnlp.839",
+    pages = "14361--14373",
+    abstract = "The lack of high-quality training data remains a significant challenge in NLP. Manual annotation methods, such as crowdsourcing, are costly, require intricate task design skills, and, if used incorrectly, may result in poor data quality. From the other hand, LLMs have demonstrated proficiency in many NLP tasks, including zero-shot and few-shot data annotation. However, they often struggle with text detoxification due to alignment constraints and fail to generate the required detoxified text. This work explores the potential of modern open source LLMs to annotate parallel data for text detoxification. Using the recent technique of activation patching, we generate a pseudo-parallel detoxification dataset based on ParaDetox. The detoxification model trained on our generated data shows comparable performance to the original dataset in automatic detoxification evaluation metrics and superior quality in manual evaluation and side-by-side comparisons."
+}
+```
 
 **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]
+Moskovskiy, D., Pletenev, S., & Panchenko, A. (2024, November). Llms to replace crowdsourcing for parallel data creation? the case of text detoxification. In Findings of the Association for Computational Linguistics: EMNLP 2024 (pp. 14361-14373).
 
 ## Model Card Contact
 
-[More Information Needed]
+[https://huggingface.co/etomoscow]