README.md

---
tags:
  - financial NLP
  - named entity recognition
  - sequence labeling
datasets:
  - AAU-NLP/hifi-kpi-lite
model_name: "Lite-BERT-SL"
library_name: "transformers"
pipeline_tag: "token-classification"
base_model: "bert-base-uncased"
task_categories:
  - token-classification
task_ids:
  - named-entity-recognition
pretty_name: "Lite-BERT-SL: Sequence Labeling for HiFi-KPI Lite"
size_categories: "10K<n<100K"
language:
  - en
dataset_name: "HiFi-KPI Lite"
model_description: |
  Lite-BERT-SL is a **BERT-based sequence labeling model** fine-tuned on **HiFi-KPI Lite**, a manually curated subset of the 
  **HiFi-KPI dataset**. This dataset contains a smaller, expert-chosen set of **financial key performance indicators (KPIs)**.

  Unlike the full HiFi-KPI dataset, HiFi-KPI Lite focuses on **four expert-mapped KPI clusters** (e.g., revenue, earnings, 
  EPS, EBIT).
  
dataset_link: "https://huggingface.co/datasets/AAU-NLP/hifi-kpi-lite"
repo_link: "https://github.com/rasmus393/HiFi-KPI"
---

## **Lite-BERT-SL**

### **Model Description**  
Lite-BERT-SL is a **BERT-based sequence labeling model** fine-tuned on the **[HiFi-KPI Lite dataset](https://huggingface.co/datasets/AAU-NLP/hifi-kpi-lite)**, 
which is a manually curated version of **HiFi-KPI** with four general KPI categories.

### **Use Cases**
- Identifying **generalized KPIs** from SEC **10-K and 10-Q** reports  
- **Financial document parsing** with entity recognition  

### **Performance**
- Trained on **HiFi-KPI Lite**, which includes a **manually curated subset** of financial KPIs For performance table see [HiFi-KPI Lite](https://huggingface.co/datasets/AAU-NLP/hifi-kpi-lite)  


### **Dataset & Code**
- **Dataset**: [HiFi-KPI Lite on Hugging Face](https://huggingface.co/datasets/AAU-NLP/hifi-kpi-lite)  
- **Code example**: [HiFi-KPI GitHub Repository](https://github.com/rasmus393/HiFi-KPI)