final model files

README.md

@@ -2,282 +2,206 @@
 base_model: Qwen/Qwen2.5-7B-Instruct
 library_name: peft
 pipeline_tag: text-generation
-license: other
-license_name: proprietary
-license_link: LICENSE
 tags:
-  - base_model:adapter:Qwen/Qwen2.5-7B-Instruct
-  - lora
-  - transformers
-  - compliance
-  - nist
-  - control-extraction
-  - regulatory
+- base_model:adapter:Qwen/Qwen2.5-7B-Instruct
+- lora
+- transformers
 ---
 
-# NIST Control Extraction LoRA Adapter
+# Model Card for Model ID
 
-A fine-tuned LoRA adapter for **Qwen2.5-7B-Instruct** specifically designed for accurate extraction of security controls from NIST framework documents. This model eliminates hallucination issues present in the base model, ensuring precise identification of controls without mistaking control enhancements or related text as valid controls.
+<!-- Provide a quick summary of what the model is/does. -->
 
-## Key Features
 
-- **Accurate Control Extraction**: Precisely identifies control IDs, titles, and descriptions from framework documents
-- **Reduced Hallucination**: Trained to distinguish between actual controls and control enhancements/related content
-- **Fast Inference**: Processes 492-page NIST documents in ~15 minutes (vs. 27 minutes with base model)
-- **Structured Output**: Returns controls in clean JSON format with `<END>` token for reliable parsing
-
----
 
 ## Model Details
 
 ### Model Description
 
-This LoRA adapter enhances the Qwen2.5-7B-Instruct model for the specialized task of extracting security controls from compliance framework documents. The adapter was trained using a custom weighted loss function that penalizes false positives more heavily than false negatives, addressing the critical requirement in compliance auditing where incorrectly identifying a control is more problematic than missing one.
+<!-- Provide a longer summary of what this model is. -->
 
-| Property | Value |
-|----------|-------|
-| **Developed by** | Rishit Sharma |
-| **Model Type** | LoRA Adapter (PEFT) |
-| **Base Model** | [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) |
-| **Language** | English |
-| **Domain** | Compliance & Regulatory Frameworks |
-| **License** | Proprietary - No use allowed without prior permission |
 
----
 
-## Model Architecture
+- **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]
 
-### LoRA Configuration
+### Model Sources [optional]
 
-| Parameter | Value |
-|-----------|-------|
-| **Rank (r)** | 16 |
-| **Alpha** | 32 |
-| **Dropout** | 0.05 |
-| **Target Modules** | `q_proj`, `k_proj`, `v_proj`, `o_proj` |
-| **Bias** | None |
-| **Task Type** | CAUSAL_LM |
+<!-- Provide the basic links for the model. -->
 
-### Quantization (Training)
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
 
-| Parameter | Value |
-|-----------|-------|
-| **Quantization** | 4-bit (QLoRA) |
-| **Quant Type** | NF4 |
-| **Double Quantization** | Enabled |
-| **Compute Dtype** | bfloat16 |
+## Uses
 
-### Special Tokens
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
 
-- **`<END>`**: Custom stop token appended to outputs for reliable generation termination
+### Direct Use
 
----
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
 
-## Intended Use
+[More Information Needed]
 
-### Primary Use Case
+### Downstream Use [optional]
 
-Building autonomous compliance auditing agents that can:
-- Parse and analyze framework documents (PDF/text)
-- Extract structured control information automatically
-- Verify deployment status of controls within an organization
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
 
-### Target Users
+[More Information Needed]
 
-- Compliance Officers & Auditors
-- GRC (Governance, Risk, Compliance) Teams
-- Security Analysts
-- Organizations undergoing NIST compliance assessments
+### Out-of-Scope Use
 
----
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
 
-## Quick Start
-
-### Installation
-
-```bash
-pip install transformers peft torch accelerate bitsandbytes
-```
-
-### Loading the Model
-
-```python
-from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
-from peft import PeftModel
-import torch
-
-# Quantization config (optional, for memory efficiency)
-bnb_config = BitsAndBytesConfig(
-    load_in_4bit=True,
-    bnb_4bit_use_double_quant=True,
-    bnb_4bit_quant_type="nf4",
-    bnb_4bit_compute_dtype=torch.bfloat16
-)
-
-# Load base model
-base_model = AutoModelForCausalLM.from_pretrained(
-    "Qwen/Qwen2.5-7B-Instruct",
-    quantization_config=bnb_config,
-    device_map="auto",
-    trust_remote_code=True
-)
-
-# Load tokenizer
-tokenizer = AutoTokenizer.from_pretrained("path/to/final_adapter")
-
-# Load LoRA adapter
-model = PeftModel.from_pretrained(base_model, "path/to/final_adapter")
-```
-
-### Inference Example
-
-```python
-system_prompt = """You are a senior Compliance Auditor and Regulatory Analyst specialized in ISO, NIST, and statutory frameworks."""
-
-messages = [
-    {"role": "system", "content": system_prompt},
-    {"role": "user", "content": f"Analyze this text:\n\n{page_text}"}
-]
-
-input_ids = tokenizer.apply_chat_template(
-    messages,
-    add_generation_prompt=True,
-    return_tensors="pt"
-).to(model.device)
-
-outputs = model.generate(
-    input_ids,
-    max_new_tokens=512,
-    temperature=0.1,
-    do_sample=False
-)
-
-response = tokenizer.decode(outputs[0], skip_special_tokens=True)
-```
-
-### Expected Output Format
-
-```json
-[
-    {
-        "control_id": "AC-1",
-        "control_title": "Access Control Policy and Procedures",
-        "control_desc": "Description of the control..."
-    }
-]
-<END>
-```
+[More Information Needed]
 
----
+## Bias, Risks, and Limitations
 
-## Training Details
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### 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.
 
-### Training Configuration
-
-| Parameter | Value |
-|-----------|-------|
-| **Hardware** | NVIDIA RTX 5070 Ti |
-| **Training Time** | ~1 hour |
-| **Epochs** | 14 (with early stopping) |
-| **Batch Size** | 1 (effective: 8 with gradient accumulation) |
-| **Learning Rate** | 1e-5 |
-| **Optimizer** | Paged AdamW 8-bit |
-| **LR Scheduler** | Cosine |
-| **Warmup Ratio** | 0.05 |
-| **Max Gradient Norm** | 0.3 |
-| **Precision** | FP16 |
+[More Information Needed]
+
+## Training Details
 
 ### Training Data
 
-- **Source**: NIST SP 800-53 Framework (492 pages)
-- **Dataset Creation**: Custom pipeline using Gemini Pro for initial extraction, followed by manual verification
-- **Data Balance**: ~60% positive samples (pages with controls), ~40% negative samples (pages without controls)
-- **Format**: JSONL with chat template structure
+<!-- 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. -->
 
-### Custom Loss Function
+[More Information Needed]
 
-A **Weighted Loss Trainer** was implemented to address the asymmetric cost of errors in compliance:
+### Training Procedure
 
-- **Positive Weight**: 2.0x (penalizes missing actual controls less than falsely identifying controls)
-- **Rationale**: In compliance auditing, falsely identifying a control (hallucination) is more problematic than missing one, as it can lead to incorrect compliance assessments
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
 
-```python
-# Samples with controls are weighted 2x during loss computation
-weights = torch.where(has_control, 2.0, 1.0)
-weighted_loss = (sample_loss * weights).mean()
-```
+#### Preprocessing [optional]
 
----
+[More Information Needed]
 
-## 📈 Evaluation & Performance
 
-| Metric | Base Qwen2.5-7B | This Adapter |
-|--------|-----------------|--------------|
-| **Processing Time (492 pages)** | ~27 minutes | ~15 minutes |
-| **Hallucination Rate** | High | Minimal |
-| **Control Enhancement Confusion** | Frequent | Resolved |
+#### 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 -->
 
-## ⚠️ Limitations & Risks
+#### Speeds, Sizes, Times [optional]
 
-### Current Limitations
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
 
-- **Framework Specificity**: Optimized primarily for NIST SP 800-53; performance on other frameworks (ISO 27001, SOC 2, etc.) may vary
-- **Language**: Trained on English documents only
-- **Document Format**: Best performance on well-structured PDF documents
+[More Information Needed]
 
-### Known Risks
+## Evaluation
 
-- May require additional fine-tuning for non-NIST frameworks
-- Performance depends on input text quality and preprocessing
-- Should be validated by human auditors for critical compliance decisions
+<!-- This section describes the evaluation protocols and provides the results. -->
 
-### Future Improvements
+### Testing Data, Factors & Metrics
 
-- [ ] Training on ISO 27001/27002 frameworks
-- [ ] Multi-framework support (SOC 2, HIPAA, PCI-DSS)
-- [ ] Improved handling of complex document layouts
-- [ ] Longer training with expanded dataset
+#### Testing Data
 
----
+<!-- This should link to a Dataset Card if possible. -->
 
-## 📄 License
+[More Information Needed]
 
-**Proprietary License** - This model is not available for public use without explicit prior permission from the developer.
+#### Factors
 
-For licensing inquiries, please contact via the channels below.
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
 
----
+[More Information Needed]
 
-## 📬 Contact
+#### Metrics
 
-| Channel | Link |
-|---------|------|
-| **Email** | [rishitshar36@gmail.com](mailto:rishitshar36@gmail.com) |
-| **GitHub** | [github.com/rishit836](https://github.com/rishit836) |
-| **Project Repository** | [control-extraction-using-llm-finetuned](https://github.com/rishit836/control-extraction-using-llm-finetuned) |
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
 
----
+[More Information Needed]
 
-## 🙏 Acknowledgments
+### Results
 
-- [Qwen Team](https://github.com/QwenLM/Qwen2.5) for the excellent base model
-- [Hugging Face](https://huggingface.co/) for the Transformers and PEFT libraries
-- NIST for the publicly available SP 800-53 framework documentation
+[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]
 
-## 📚 Citation
+## Model Card Contact
 
-If you use this model in your research or project, please cite:
+[More Information Needed]
+### Framework versions
 
-```bibtex
-@misc{sharma2026nist-control-extraction,
-    title={NIST Control Extraction LoRA Adapter for Qwen2.5-7B},
-    author={Sharma, Rishit},
-    year={2026},
-    publisher={GitHub},
-    howpublished={\url{https://github.com/rishit836/control-extraction-using-llm-finetuned}}
-}
-```
+- PEFT 0.18.1

adapter_config.json

@@ -29,8 +29,8 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
     "o_proj",
+    "v_proj",
     "k_proj",
     "q_proj"
   ],

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2bf4dc666d62aa595c65f300110b7411d50eeeb252e371ab8c6ab6f6a8d54c8a
+oid sha256:83af2cf6ccdcbe71f86aab5fd5034fcba4e1b5373f3b57a0a32b7f1b63b195a5
 size 4388968992