# High-precision Classification

Multilingual E5 large instruction model, supporting text embedding and classification tasks in multiple languages with strong cross-language capabilities.

A prompt injection detection model fine-tuned based on microsoft/mdeberta-v3-base, trained with multiple datasets to identify malicious prompt injection attacks.

Linq-Embed-Mistral is an embedding model based on the Mistral architecture, specializing in text classification, retrieval, and clustering tasks, with outstanding performance across multiple MTEB benchmarks.

This is a BERT-based sentiment analysis model for Indonesian language, capable of classifying text into positive or negative sentiments.

A fine-tuned AI-generated text detection model based on DeBERTa-v3-large, optimized for academic scenarios

An EfficientNetB4 image classification model fine-tuned on the Food101 dataset, capable of recognizing 101 food categories

A Vision Transformer (ViT)-based image classification model for distinguishing AI-generated from human-created images, achieving 98% accuracy.

A BERT-LoRA fine-tuned model for efficient detection of malicious URLs, including phishing, malware, and tampering threats.

A zero-shot image classification model based on the open_clip library, specializing in medical field image analysis

An image classification solution implemented using Hugging Face's pre-trained Vision Transformer (ViT) model, capable of classifying images into six types of garbage

This model is a facial expression classifier fine-tuned on the FER2013 dataset based on Google's ViT base model, capable of classifying images into four facial expressions.

A diabetic retinopathy classification model based on Vision Transformer (ViT) architecture for automatic analysis of fundus images

This model is a text classification model fine-tuned based on RoBERTa, specializing in identifying whether user input text belongs to medical consultation or casual conversation categories.

A phishing email detection model fine-tuned from Llama-Guard-3-1B, capable of efficiently identifying phishing attacks.

This project utilizes the Whisper model for speech emotion recognition, capable of classifying audio into different emotional categories such as happiness, sadness, and surprise.

An image classification model fine-tuned on the BrainMRI dataset based on Google's ViT base model

Anime-style AI-generated image detection model based on Vision Transformer architecture, achieving an accuracy of up to 94.68%

This is a PyTorch-based text classification model suitable for various text classification tasks.

A Russian spam detection model based on the Deberta architecture, capable of classifying text as spam or non-spam.

An image classification model based on PyTorch and HuggingPics, specifically designed to recognize different types of rooms.

This is a merged model composed of multiple foundation models including bert-large-uncased, WhereIsAI/UAE-Large-V1, BAAI/bge-large-en-v1.5, mixedbread-ai/mxbai-embed-large-v1, and avsolatorio/GIST-large-Embedding-v0. It is primarily used for text classification, retrieval, and clustering tasks.

This model is a fine-tuned audio classification model based on asapp/sew-mid-100k on the alexandreacff/kaggle-fake-detection dataset, designed for fake audio detection.

Audio classification model fine-tuned based on facebook/wav2vec2-large for detecting forged audio

Venusaur is a sentence embedding model developed based on the Mihaiii/Bulbasaur foundation model, focusing on sentence similarity and feature extraction tasks.

A 7B-parameter sentence embedding model based on the Qwen1.5 architecture, focusing on sentence similarity calculation and multi-task evaluation

A deepfake image detection model fine-tuned based on the SigLIP2 vision-language encoder for binary classification of real vs. synthetic images

A deepfake audio detection model fine-tuned based on Microsoft's WavLM-base, achieving excellent performance on the ASVspoof 2019 dataset with an accuracy of 99.79%

Multilingual toxicity classifier based on xlm-roberta-large architecture, supporting toxicity detection in 9 languages

Financial news sentiment analysis model fine-tuned on DeBERTa-v3-small, achieving an F1 score of 0.9940 on the evaluation set

A fashion collection classification model fine-tuned based on Google Vision Transformer (ViT), capable of recognizing clothing collections from 15 top fashion brands.

A sign language classification model fine-tuned based on Google Vision Transformer (ViT), achieving an accuracy of 80.56%

A fine-tuned audio emotion classification model based on facebook/wav2vec2-base for analyzing emotional states in speech

An image classification model based on Vision Transformer architecture, used to detect real images versus AI-generated fake images.

This model uses Google's ViT-base architecture to predict the presence of beards in facial images, achieving 100% accuracy on the test set.

A CNN model for classifying Arabic handwritten characters (Hijaiyah letters), trained on an Arabic handwritten character dataset with an accuracy of 97.64%.

Text classification model based on ELECTRA architecture for detecting expressions of suicidal tendencies in text

This model is used for audio classification tasks and can classify and recognize audio data.

An image classification model fine-tuned based on ResNet-50, achieving 96.45% accuracy in landmark recognition tasks

GTE-small is a compact general-purpose text embedding model suitable for various natural language processing tasks, including sentence similarity calculation, text classification, and retrieval.

This is a Japanese sentiment analysis model based on the BERT architecture, fine-tuned on the JGLUE MARC-ja dataset.