# Multi-domain Adaptation

DeepSeek-R1T-Chimera is an open-source weights model that combines the intelligence of DeepSeek-R1 with the token efficiency of DeepSeek-V3.

This is a fine-tuned BERT-based model for English quality-related text classification tasks, part of the Multi2ConvAI project.

PEG is a model that achieves robust text retrieval through progressive learning, adjusting loss weights based on the difficulty levels of negative samples.

Turkish pre-trained model based on ModernBERT architecture, supporting 8192 context length with excellent performance across multiple domains

A large language model based on 12B parameters, supporting text generation tasks, released under the Apache-2.0 license.

DeepSeek-R1-Distill-Llama-8B is an 8B-parameter large language model based on the Llama architecture, optimized through distillation training for text generation tasks.

A Thai sentiment analysis model fine-tuned based on the WangchanBERTa model, used to analyze the emotional tendency of Thai text.

Hugging Face Transformers is a library providing pre-trained deep learning models, supporting tasks in natural language processing, computer vision, and other domains.

An English-Korean translation model based on exaone-3-7.8B-it training, focused on translation tasks of instruction datasets

A Persian Q&A specialized version fine-tuned based on Llama3, capable of accurately answering Persian questions in context

This project is an open-source project following the MIT license and has certain open-source value.

An English-Korean translation model trained on Llama-3-8B-it, specifically designed for translating English instruction datasets

A text classification model trained with AutoTrain, suitable for few-shot learning scenarios, capable of efficiently classifying news articles.

GNER-T5-large is a generative named entity recognition model based on the Flan-T5-large architecture, focusing on improving zero-shot recognition capabilities in unseen entity domains.

A Flan-T5-Base fine-tuned intent classification model for categorizing customer queries into predefined categories

TinyLlama is a compact language model with 1.1 billion parameters, adopting the same architecture and tokenizer as Llama 2, suitable for applications with limited computational and memory resources.

TinyLlama is a small language model with 1.1 billion parameters, adopting the same architecture and tokenizer as Llama 2, suitable for resource-constrained application scenarios.

GNER-T5-xl is a generative named entity recognition model based on Flan-T5-xl, significantly enhancing zero-shot recognition capabilities through negative instance training

Parakeet TDT 1.1B is an automatic speech recognition (ASR) model jointly developed by NVIDIA NeMo and Suno.ai, capable of transcribing speech into lowercase English letters.

A text embedding model developed by Salesforce Research, trained on E5-mistral-7b-instruct and Mistral-7B-v0.1, primarily used for text retrieval tasks.

Parakeet CTC 0.6B is an automatic speech recognition model jointly developed by NVIDIA NeMo and Suno.ai, based on the FastConformer architecture with approximately 600 million parameters, supporting English speech transcription.

Parakeet RNNT 0.6B is an automatic speech recognition model jointly developed by NVIDIA NeMo and Suno.ai, based on the FastConformer architecture with approximately 600 million parameters, specifically designed for transcribing English speech into text.

A Chinese speech recognition model fine-tuned based on whisper-large-v2, achieving a 30-70% relative performance improvement in multiple Chinese speech recognition benchmarks.

RobBERT-2023 is a Dutch language model based on the RoBERTa architecture, developed by KU Leuven, Ghent University, and TU Berlin, and is one of the state-of-the-art language models for Dutch.

Sheared-LLaMA-2.7B is a lightweight language model derived from Llama-2-7b through pruning and continued pretraining, consuming only a 50B token budget.

This is a sentence encoder model based on sentence-transformers, capable of converting text into 768-dimensional vector representations, suitable for tasks such as semantic search and sentence similarity calculation.

GTE-small is a general text embedding model trained by Alibaba DAMO Academy, based on the BERT framework, suitable for tasks such as information retrieval and semantic text similarity.

A Thai text-to-speech model based on the Tacotron2 architecture, trained using a modified Common Voice Thai dataset

A Russian text and dialogue summarization model fine-tuned based on ruT5-base, supporting multi-domain Russian text summarization tasks

Korean speech recognition model based on the Whisper Small architecture, fine-tuned on multi-domain Korean datasets

A sentence transformer model optimized for phrases, mapping phrases into a 768-dimensional dense vector space, suitable for tasks like clustering or semantic search.

An English long-text summarization model fine-tuned based on Pegasus-x-large, supporting abstractive summarization of complex texts such as academic manuscripts and meeting minutes

EurekaQA is an AI Q&A model based on advanced machine learning algorithms, capable of automatically extracting information to answer questions by analyzing text data.

A Chinese PEGASUS-large model specialized in text summarization tasks, fine-tuned on multiple Chinese summarization datasets

JASMINE is a series of Arabic GPT models designed for few-shot learning, with parameters ranging from 300 million to 6.7 billion, pretrained on 235GB of text data.

A specialized Chinese version of PAGASUS-base optimized for Chinese text summarization tasks, fine-tuned on multiple Chinese summarization datasets

A Chinese PEGASUS-large model specialized in text summarization tasks, fine-tuned on multiple Chinese summarization datasets

This is a neural machine translation model for translating from Hungarian to English, part of the OPUS-MT project.

This is a large-scale neural machine translation model based on the Transformer architecture, specifically designed for translating English to Romanian.

This is a neural machine translation model for English to Latvian translation, part of the OPUS-MT project, using transformer-big architecture.