# Fine-tuning optimization

Zk1-QA-VN-test2 is a Vietnamese Q&A model fine-tuned based on ZycckZ/Zk1-QA-VN-test, which performs excellently in Q&A tasks.

An image segmentation model fine-tuned based on nvidia/mit-b3, trained on the NICOPOI-9/morphpad_coord_hgo_512_4class dataset, excelling in high-precision image segmentation tasks.

A lightweight image segmentation model based on the SegFormer architecture, fine-tuned on the NICOPOI-9/morphpad_coord_hgo_512_4class dataset, specifically designed for high-precision semantic segmentation tasks.

A lightweight image segmentation model based on the SegFormer architecture, fine-tuned on specific datasets, excelling in high-precision pixel-level classification tasks

LlaMaestra is a fine-tuned small Llama model focused on text translation between English and Italian.

Based on the transformers library, this is a fine-tuned version of openai/whisper-large-v3-turbo on the common_voice_11_0 dataset, optimized specifically for Arabic speech recognition.

An open-source generative text model fine-tuned based on Deepseek-coder-1.3B, focusing on Kotlin code generation tasks

The 4-bit quantization version of Phi-3-mini-4k-instruct, quantized using the bitsandbytes tool, is designed specifically for fine-tuning.

This is a 7B-parameter instruction fine-tuned model for Polish language, developed based on SpeakLeash's Bielik-7B-v0.1 model and quantized in GGUF format.

A speech recognition model fine-tuned on the Common Voice 16.0 dataset based on OpenAI Whisper Large V3

This model achieves selective forgetting of 'Harry Potter' series knowledge from large language models through fine-tuning while maintaining other performance metrics.

This model is an image captioning model based on the BLIP architecture, specifically fine-tuned for high-level scene descriptions.

A voice processing model fine-tuned based on microsoft/wavlm-large, achieving 80% accuracy and 79.57% F1 score on the evaluation set

A Romanian to English translation model fine-tuned based on Helsinki-NLP/opus-mt-roa-en, trained on 2 million records.

Fine-tuned model based on Swin Transformer architecture for AI-generated content detection tasks

A speech recognition model fine-tuned on the Common Voice 11.0 German dataset based on openai/whisper-small, supporting German speech-to-text with case and punctuation prediction.

This model is a fine-tuned Turkish speech recognition model based on OpenAI's Whisper Medium, with significantly reduced word error rate.

A fine-tuned version based on the DistilBERT base model, suitable for text classification tasks

Hate speech/offensive language detection model fine-tuned based on DistilBERT-base-uncased