I

Intermediate Base

Developed by funnel-transformer
A Transformer model pre-trained on English corpus using ELECTRA-like self-supervised approach, suitable for sentence summarization tasks.
Downloads 20
Release Time : 3/2/2022

Model Overview

Funnel Transformer is a self-supervised pre-trained Transformer model on large-scale English corpus, primarily used for extracting text features for downstream task fine-tuning.

Model Features

Self-supervised Pretraining
Trained using ELECTRA-like self-supervised approach without manual annotation data
Efficient Sequence Processing
Output hidden state sequence length is one-fourth of input, suitable for sentence summarization tasks
Case Insensitive
Model is case-insensitive, treating 'english' and 'English' as identical

Model Capabilities

Text Feature Extraction
Sentence Classification
Sequence Classification
Token Classification
Question Answering

Use Cases

Natural Language Processing
Sentence Classification
Using model-extracted features as input to train standard classifiers
Question Answering System
Building QA systems using text features extracted by the model
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase