# Efficient Transformer

HiT-SR is a general strategy for enhancing transformer-based super-resolution methods, improving existing model structures to achieve better super-resolution performance while reducing computational burden.

Advanced model for detecting end-to-end AI-generated songs, particularly skilled at capturing long-term audio features

GLiNER-BioMed is an efficient open NER model suite based on the GLiNER framework, specifically designed for the biomedical domain to recognize various types of biomedical entities.

GLiNER-BioMed is an efficient open biomedical named entity recognition model suite based on the GLiNER framework, specifically designed for the biomedical domain, capable of recognizing multiple entity types.

GLiNER-BioMed is a specialized and efficient open biomedical NER model suite based on the GLiNER framework, achieving state-of-the-art zero-shot and few-shot performance in biomedical entity recognition tasks.

Janus-Pro is a novel autoregressive framework that unifies multimodal understanding and generation capabilities. By decoupling visual encoding paths, it uses a single Transformer architecture to handle multimodal tasks.

Deepseek LLM 7B Base is a 7B-parameter foundational large language model optimized for inference efficiency using AWQ quantization technology.

CED-Tiny is a simple audio tagging model based on ViT-Transformer, achieving state-of-the-art performance on Audioset.

CED is a simple audio tagging model based on ViT-Transformer, achieving state-of-the-art performance on Audioset.

LongT5 is a Transformer-based text-to-text model specifically designed to handle long sequence inputs, supporting up to 16,384 tokens.

A long-sequence processing model based on XLM-R extension, supporting sequences up to 4096 tokens, suitable for multilingual tasks

ELECTRA is an efficient self-supervised language representation learning method that replaces traditional generative pretraining with discriminative pretraining, significantly improving computational efficiency.

This model is a fine-tuned version of google/fnet-base on the GLUE QQP dataset for text classification tasks, specifically targeting the problem of detecting duplicate Quora question pairs.

A text classification model based on Google's FNet architecture fine-tuned on the SST-2 sentiment analysis dataset

BigBird is a Transformer model based on sparse attention, capable of processing sequences up to 4096 tokens long, suitable for long document tasks.

This is a compact version of distilbert-base-multilingual-cased, supporting English, Spanish, and Portuguese, while maintaining the original model's accuracy.

Extended XLM-RoBERTa model supporting sequences up to 4096 tokens, suitable for multilingual tasks

ALBERT XXLarge v1 is a Transformer model pretrained on English corpus using Masked Language Modeling (MLM) objective with parameter-sharing features.