# Zero-shot classification

This is a transformers model for zero-shot classification tasks, which can classify text without a large amount of labeled data.

A vision-language model based on the OpenCLIP-ViT-B-16 architecture. By introducing test-time registers to optimize the internal representation, it solves the problem of feature map artifacts.

An instruction fine-tuned model optimized based on the Mistral architecture, suitable for zero-shot classification tasks

FG-CLIP is a fine-grained vision and text alignment model that achieves global and region-level image-text alignment through two-stage training, enhancing fine-grained visual understanding ability.

A zero-shot classification model fine-tuned from MoritzLaurer/deberta-v3-base-zeroshot-v2.0-c using the SmartShot method with an NLI framework

A zero-shot classification model fine-tuned based on roberta-base-zeroshot-v2.0-c, enhanced with SmartShot method and synthetic data

An emotion detection model fine-tuned based on mDeBERTa-v3, supporting emotion classification in Indonesian and English

A natural language inference model optimized through multi-task fine-tuning based on the ModernBERT-large model, excelling in zero-shot classification and NLI tasks.

A zero-shot classifier fine-tuned based on ModernBERT-base, efficient and fast with low memory usage, suitable for various text classification tasks.

A zero-shot classifier fine-tuned based on ModernBERT-large, efficient and fast with low memory usage, suitable for various text classification tasks.

A Japanese CLIP model trained based on the OpenCLIP framework, trained on a dataset of 1.45 billion Japanese image-text pairs, supporting zero-shot image classification and image-text retrieval tasks

CLIP model image encoder based on ResNet50 architecture with 64x width expansion, using Global Average Pooling (GAP) strategy

A ResNet50x16 variant model based on the CLIP framework, focused on image feature extraction

ResNet50x4 variant model based on the CLIP framework, designed for image feature extraction

A ResNet50 variant based on the visual encoder part of the CLIP model, extracting image features through Global Average Pooling (GAP)

CLIP-style image encoder based on ResNet50 architecture, trained on CC12M dataset, extracting features through Global Average Pooling (GAP)

A vision transformer model based on the CLIP architecture, focused on image feature extraction, released by Apple.

A natural language inference model fine-tuned based on ModernBERT-large, specifically designed for zero-shot classification tasks

ViT-Huge visual encoder based on the CLIP framework, trained on the laion2B dataset, supports image feature extraction

Vision Transformer model based on CLIP architecture, designed for image feature extraction, trained on the laion2B dataset

ConvNeXt Base model based on the CLIP framework, trained on the LAION-Aesthetic dataset, suitable for image feature extraction tasks.

ModernBERT is a model fine-tuned on multi-task natural language inference (NLI) tasks, excelling in zero-shot classification and long-context reasoning.

Japanese CLIP model trained on OpenCLIP framework, supporting zero-shot image classification tasks

A zero-shot text classification model fine-tuned on DeBERTa-v3-base, suitable for scenarios with scarce labeled data or rapid prototyping.

LLM2CLIP is an innovative approach that leverages large language models (LLMs) to unlock the potential of CLIP. It enhances text discriminability through a contrastive learning framework, breaking the limitations of the original CLIP text encoder.

LLM2CLIP is an innovative method that leverages large language models (LLMs) to extend CLIP's capabilities, enhancing text discriminability through a contrastive learning framework and significantly improving cross-modal task performance.

This is the OpenVINO optimized version of the facebook/bart-large-mnli model for zero-shot text classification tasks.

A dual-framework compatible vision model trained on the MetaCLIP-2.5B dataset, supporting both OpenCLIP and timm frameworks

ResNet50 model trained on the YFCC-15M dataset, compatible with both open_clip and timm frameworks, supporting zero-shot image classification tasks.

A compact zero-shot classification model based on DeBERTa architecture, optimized for long-text natural language inference tasks, converted to ONNX format for web compatibility

EXLMR is an extended version of XLM-R that supports new languages by expanding the tokenizer vocabulary to mitigate out-of-vocabulary issues, specifically optimized for low-resource Ethiopian languages.

Marqo-FashionSigLIP is a multimodal embedding model optimized for fashion product search, with a 57% improvement in MRR and recall rate compared to FashionCLIP.

Quantized ONNX model based on DeBERTa-v3-base, suitable for zero-shot text classification tasks

Safe-CLIP is an enhanced vision-language model designed to mitigate risks associated with Not Safe For Work (NSFW) content in AI applications.

An efficient zero-shot classifier trained on synthetic data, suitable for topic classification, sentiment analysis, and reranking tasks in RAG workflows.

GLiClass is an efficient zero-shot classifier inspired by GLiNER, suitable for text classification, sentiment analysis, and reranking tasks in RAG workflows.

An efficient zero-shot classifier trained on synthetic data, suitable for topic classification, sentiment analysis, and reranking tasks in RAG workflows

GLiClass is an efficient zero-shot classifier trained on synthetic data, suitable for text classification, sentiment analysis, and reranking tasks in RAG workflows.

A zero-shot classification model based on the Transformers library, capable of performing classification tasks without task-specific training data.

Based on the DeBERTa-v3-base model, the context length is extended to 1280, and fine-tuned for 250,000 steps on the tasksource dataset, focusing on natural language inference and zero-shot classification tasks.