Lite - BERT - SL Open - Source Model - Free for Financial Document Parsing and Key Performance Indicator Identification and Extraction

Lite BERT SL

Developed by AAU-NLP

Lite-BERT-SL is a sequence labeling model based on BERT, focusing on the identification of key performance indicators (KPIs) in the financial domain, suitable for financial document parsing and information extraction.

Sequence Labeling

Transformers

English#Financial KPI Identification #SEC Document Parsing #BERT Fine-tuning

Downloads 405

Release Time : 2/19/2025

Model Overview

This model has been fine-tuned on the HiFi-KPI Lite dataset and can effectively identify financial key performance indicators, providing support for financial document parsing.

Model Features

Professional Domain Adaptation

Optimized for the identification of key performance indicators in the financial domain, suitable for processing financial-related documents.

Dataset Refinement

Trained on the HiFi-KPI Lite dataset, which is a manually selected subset of HiFi-KPI, focusing on four expert-mapped KPI clusters.

Multi-scenario Application

Can be used to identify general KPIs from SEC's 10-K and 10-Q reports, as well as perform entity recognition and parsing of financial documents.

Model Capabilities

Financial Key Performance Indicator Identification

Financial Document Parsing

Named Entity Recognition

Use Cases

Financial Document Processing

General KPI Identification

Identify general KPIs from the 10-K and 10-Q reports of the U.S. Securities and Exchange Commission (SEC).

Financial Document Parsing

Parse financial documents through entity recognition.

🚀 Lite-BERT-SL: Sequence Labeling for HiFi-KPI Lite

Lite-BERT-SL is a BERT-based sequence labeling model. It is fine-tuned on the HiFi-KPI Lite dataset, which is a manually curated subset of the HiFi-KPI dataset. This model is designed to handle financial named entity recognition and sequence labeling tasks, offering valuable insights for financial document analysis.

🚀 Quick Start

This section provides a high - level overview of the Lite - BERT - SL model. For detailed usage, please refer to the code example in the HiFi - KPI GitHub Repository.

✨ Features

Fine - Tuned on Specific Dataset: Lite-BERT-SL is fine-tuned on the HiFi-KPI Lite dataset, a manually curated version of HiFi - KPI with four general KPI categories.
Useful for Financial Tasks: It can be used for identifying generalized KPIs from SEC 10 - K and 10 - Q reports and financial document parsing with entity recognition.

📚 Documentation

Model Description

Lite-BERT-SL is a BERT-based sequence labeling model fine-tuned on the HiFi-KPI Lite dataset, 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

Dataset & Code

Dataset: HiFi-KPI Lite on Hugging Face
Code example: HiFi-KPI GitHub Repository

📄 License

No license information is provided in the original document, so this section is skipped.

Property	Details
Model Type	BERT-based sequence labeling model
Training Data	HiFi-KPI Lite, a manually curated subset of the HiFi-KPI dataset, focusing on four expert-mapped KPI clusters (e.g., revenue, earnings, EPS, EBIT)
Dataset Link	HiFi-KPI Lite
Repo Link	HiFi-KPI GitHub Repository
Pipeline Tag	token-classification
Base Model	bert-base-uncased
Task Categories	token-classification
Task IDs	named-entity-recognition
Size Categories	10K < n < 100K
Language	en
Dataset Name	HiFi-KPI Lite

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご