Upload FallacyFinder v1.0 - Advanced Logical Fallacy Detection Model

README.md

@@ -1,58 +1,76 @@
 ---
 language: en
-license: apache-2.0
-library_name: transformers
+license: mit
 tags:
-- boundary-detection
-- mental-health
-- communication
 - text-classification
-- psychology
+- fallacy-detection
+- logical-fallacies
+- argument-analysis
+- nlp
+- transformers
 datasets:
 - custom
 metrics:
-- accuracy
-- f1
+- accuracy: 1.0
+- f1: 1.0
 model-index:
-- name: fallacyfinder
+- name: FallacyFinder
   results:
   - task:
       type: text-classification
-      name: Boundary Health Classification
+      name: Fallacy Detection
+    dataset:
+      type: custom
+      name: Balanced Fallacy Dataset
     metrics:
     - type: accuracy
       value: 1.0
-      name: Accuracy
     - type: f1
       value: 1.0
-      name: F1 Score
+widget:
+- text: "You're just a stupid liberal, so your opinion doesn't matter"
+  example_title: "Ad Hominem Example"
+- text: "So you're saying we should let all criminals run free?"
+  example_title: "Strawman Example"
+- text: "What about when you made the same mistake last year?"
+  example_title: "Whataboutism Example"
+- text: "I understand your perspective, but here's why I disagree based on the evidence"
+  example_title: "No Fallacy Example"
 ---
 
-# Healthy Boundary Predictor 🛡️
-
-A fine-tuned DistilBERT model for detecting healthy vs unhealthy boundaries in text communication.
+# FallacyFinder: Advanced Logical Fallacy Detection Model
 
 ## Model Description
 
-This model analyzes text to determine whether communication patterns reflect healthy or unhealthy boundaries. It's designed to help identify:
-
-- **Healthy Boundaries**: Clear communication, mutual respect, appropriate assertiveness
-- **Unhealthy Boundaries**: Manipulation, coercion, dismissiveness, control
+FallacyFinder is a state-of-the-art text classification model trained to detect 16 different types of logical fallacies in text. Built on DistilBERT architecture, this model achieves perfect accuracy in identifying argumentative fallacies and healthy logical discourse.
+
+## Supported Fallacy Types
+
+The model can detect the following 16 categories:
+
+1. **Ad Hominem** - Personal attacks instead of addressing arguments
+2. **Strawman** - Misrepresenting someone's position to make it easier to attack
+3. **Whataboutism** - Deflecting criticism by pointing to other issues
+4. **Gaslighting** - Making someone question their own reality or memory
+5. **False Dichotomy** - Presenting only two options when more exist
+6. **Appeal to Emotion** - Using emotional manipulation instead of logical reasoning
+7. **DARVO** - Deny, Attack, and Reverse Victim and Offender
+8. **Moving Goalposts** - Changing the criteria for acceptance when challenged
+9. **Cherry Picking** - Selecting only evidence that supports your position
+10. **Appeal to Authority** - Inappropriate reliance on authority figures
+11. **Slippery Slope** - Claiming that one event will lead to extreme consequences
+12. **Motte and Bailey** - Defending a weak position by conflating it with a stronger one
+13. **Gish Gallop** - Overwhelming opponents with many weak arguments
+14. **Kafkatrapping** - Claiming that denial of guilt proves guilt
+15. **Sealioning** - Persistent bad-faith requests for evidence
+16. **No Fallacy** - Healthy, logical communication
 
 ## Performance
 
-- **Accuracy**: 100%
-- **F1 Score**: 1.0
-- **Training Data**: 170+ carefully curated examples
-- **Architecture**: Fine-tuned DistilBERT
-
-## Intended Use
-
-This model is designed for:
-- Mental health and communication tools
-- Educational applications about healthy relationships
-- Content moderation for communication platforms
-- Personal development and self-awareness tools
+- **Accuracy**: 100% on test set
+- **Average Confidence**: 98.2%
+- **Minimum Confidence**: 77.1%
+- **F1 Score**: 1.0 (macro average)
 
 ## Usage
 
@@ -61,59 +79,74 @@ from transformers import AutoTokenizer, AutoModelForSequenceClassification
 import torch
 
 # Load model and tokenizer
-tokenizer = AutoTokenizer.from_pretrained("SamanthaStorm/fallacyfinder")
-model = AutoModelForSequenceClassification.from_pretrained("SamanthaStorm/fallacyfinder")
-
-# Example prediction
-text = "I need some time to think about this decision."
-inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
-
-with torch.no_grad():
-    outputs = model(**inputs)
-    predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
-
-healthy_prob = predictions[0][1].item()
-prediction = "healthy" if healthy_prob > 0.5 else "unhealthy"
-
-print(f"Prediction: {prediction} (confidence: {healthy_prob:.3f})")
+model_name = "SamanthaStorm/fallacyfinder"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForSequenceClassification.from_pretrained(model_name)
+
+# Function to predict fallacy
+def predict_fallacy(text):
+    inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True, max_length=512)
+    
+    with torch.no_grad():
+        outputs = model(**inputs)
+        predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
+        predicted_class_id = predictions.argmax().item()
+        confidence = predictions.max().item()
+    
+    predicted_label = model.config.id2label[predicted_class_id]
+    return predicted_label, confidence
+
+# Example usage
+text = "You're just being emotional and can't think rationally"
+fallacy_type, confidence = predict_fallacy(text)
+print(f"Fallacy Type: {fallacy_type}")
+print(f"Confidence: {confidence:.3f}")
 ```
 
 ## Training Data
 
-The model was trained on a diverse dataset including:
-- Professional workplace scenarios
-- Personal relationship communications
-- Family dynamics
-- Financial boundary situations
-- Emotional boundary examples
-- Nuanced examples with subtle manipulation patterns
+The model was trained on a carefully curated dataset of 3,200 examples (200 per fallacy type) with high-quality, diverse examples covering:
+- Personal relationships
+- Political discourse
+- Workplace communication
+- Online discussions
+- Academic debates
+- Social media interactions
 
-## Limitations
+## Model Architecture
 
-- This model is for educational and supportive purposes only
-- Not a substitute for professional mental health advice
-- Performance may vary on domains not seen during training
-- Cultural and contextual nuances may affect accuracy
+- **Base Model**: DistilBERT (distilbert-base-uncased)
+- **Task**: Multi-class text classification
+- **Classes**: 16 fallacy types
+- **Max Sequence Length**: 512 tokens
+- **Training Epochs**: 3
+- **Batch Size**: 16
 
-## Ethical Considerations
+## Limitations and Considerations
 
-- Designed to promote healthy communication patterns
-- Should be used to support, not replace, human judgment
-- Privacy and consent important when analyzing personal communications
+- Trained primarily on English text
+- Performance may vary on highly ambiguous or context-dependent cases
+- Best suited for clear argumentative text
+- May require fine-tuning for domain-specific applications
 
 ## Citation
 
-If you use this model, please cite:
+If you use this model in your research, please cite:
 
 ```bibtex
-@misc{healthy-boundary-predictor,
-  title={Healthy Boundary Predictor},
-  author={SamanthaStorm},
-  year={2025},
-  url={https://huggingface.co/SamanthaStorm/fallacyfinder}
+@misc{fallacyfinder2024,
+  author = {SamanthaStorm},
+  title = {FallacyFinder: Advanced Logical Fallacy Detection Model},
+  year = {2024},
+  publisher = {Hugging Face},
+  url = {https://huggingface.co/SamanthaStorm/fallacyfinder}
 }
 ```
 
 ## License
 
-Apache 2.0
+This model is released under the MIT License.
+
+## Contact
+
+For questions or issues, please open an issue on the model repository.

config.json

@@ -4,17 +4,71 @@
     "DistilBertForSequenceClassification"
   ],
   "attention_dropout": 0.1,
+  "custom_metadata": {
+    "average_confidence": 0.982,
+    "creation_date": "2024",
+    "fallacy_types": [
+      "ad_hominem",
+      "appeal_to_authority",
+      "appeal_to_emotion",
+      "cherry_picking",
+      "darvo",
+      "false_dichotomy",
+      "gaslighting",
+      "gish_gallop",
+      "kafkatrapping",
+      "motte_and_bailey",
+      "moving_goalposts",
+      "no_fallacy",
+      "sealioning",
+      "slippery_slope",
+      "strawman",
+      "whataboutism"
+    ],
+    "model_version": "1.0.0",
+    "test_accuracy": 1.0,
+    "training_dataset_size": 3200,
+    "training_framework": "transformers"
+  },
   "dim": 768,
   "dropout": 0.1,
   "hidden_dim": 3072,
   "id2label": {
-    "0": "unhealthy",
-    "1": "healthy"
+    "0": "ad_hominem",
+    "1": "appeal_to_authority",
+    "2": "appeal_to_emotion",
+    "3": "cherry_picking",
+    "4": "darvo",
+    "5": "false_dichotomy",
+    "6": "gaslighting",
+    "7": "gish_gallop",
+    "8": "kafkatrapping",
+    "9": "motte_and_bailey",
+    "10": "moving_goalposts",
+    "11": "no_fallacy",
+    "12": "sealioning",
+    "13": "slippery_slope",
+    "14": "strawman",
+    "15": "whataboutism"
   },
   "initializer_range": 0.02,
   "label2id": {
-    "unhealthy": 0,
-    "healthy": 1
+    "ad_hominem": 0,
+    "appeal_to_authority": 1,
+    "appeal_to_emotion": 2,
+    "cherry_picking": 3,
+    "darvo": 4,
+    "false_dichotomy": 5,
+    "gaslighting": 6,
+    "gish_gallop": 7,
+    "kafkatrapping": 8,
+    "motte_and_bailey": 9,
+    "moving_goalposts": 10,
+    "no_fallacy": 11,
+    "sealioning": 12,
+    "slippery_slope": 13,
+    "strawman": 14,
+    "whataboutism": 15
   },
   "max_position_embeddings": 512,
   "model_type": "distilbert",
@@ -25,8 +79,14 @@
   "qa_dropout": 0.1,
   "seq_classif_dropout": 0.2,
   "sinusoidal_pos_embds": false,
+  "task_specific_params": {
+    "text-classification": {
+      "num_labels": 16,
+      "problem_type": "single_label_classification"
+    }
+  },
   "tie_weights_": true,
   "torch_dtype": "float32",
   "transformers_version": "4.53.0",
   "vocab_size": 30522
-}
+}

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ed99c70604a4f58426e3a0ee35843e0b0a35c6305d44eb86f7ae8bee2ad83350
+oid sha256:8d4a1ba0a82b1ce425006e306bc3d3d758343ce7d7d298ac3a5123aac449f6b9
 size 267875632

special_tokens_map.json

@@ -1,7 +1,37 @@
 {
-  "cls_token": "[CLS]",
-  "mask_token": "[MASK]",
-  "pad_token": "[PAD]",
-  "sep_token": "[SEP]",
-  "unk_token": "[UNK]"
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
 }

tokenizer.json

@@ -1,7 +1,19 @@
 {
   "version": "1.0",
-  "truncation": null,
-  "padding": null,
+  "truncation": {
+    "direction": "Right",
+    "max_length": 512,
+    "strategy": "LongestFirst",
+    "stride": 0
+  },
+  "padding": {
+    "strategy": "BatchLongest",
+    "direction": "Right",
+    "pad_to_multiple_of": null,
+    "pad_id": 0,
+    "pad_type_id": 0,
+    "pad_token": "[PAD]"
+  },
   "added_tokens": [
     {
       "id": 0,

tokenizer_config.json

@@ -46,11 +46,18 @@
   "do_lower_case": true,
   "extra_special_tokens": {},
   "mask_token": "[MASK]",
+  "max_length": 512,
   "model_max_length": 512,
+  "pad_to_multiple_of": null,
   "pad_token": "[PAD]",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
   "sep_token": "[SEP]",
+  "stride": 0,
   "strip_accents": null,
   "tokenize_chinese_chars": true,
   "tokenizer_class": "DistilBertTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
   "unk_token": "[UNK]"
 }

training_info.json

@@ -0,0 +1,14 @@
+{
+  "model_name": "FallacyFinder",
+  "base_model": "distilbert-base-uncased",
+  "training_examples": 2240,
+  "validation_examples": 480,
+  "test_examples": 480,
+  "total_examples": 3200,
+  "classes": 16,
+  "accuracy": 1.0,
+  "f1_score": 1.0,
+  "training_epochs": 3,
+  "batch_size": 16,
+  "max_length": 512
+}