What is supervised fine-tuning? — Klu
![](https://klu.ai/_next/image?url=%2F_next%2Fstatic%2Fmedia%2Fklu-supervised-fine-tuning-sft.ca014122.png&w=3840&q=100)
Supervised fine-tuning (SFT) is a method used in machine learning to improve the performance of a pre-trained model. The model is initially trained on a large dataset, then fine-tuned on a smaller, specific dataset. This allows the model to maintain the general knowledge learned from the large dataset while adapting to the specific characteristics of the smaller dataset.
![](https://i.ytimg.com/vi/3xV_nNl6DCI/maxresdefault.jpg)
Lecture 8: How ChatGPT Works Part 1 - Supervised Fine-Tuning
![](https://ars.els-cdn.com/content/image/1-s2.0-S266638992300288X-gr2.jpg)
Enhancing phenotype recognition in clinical notes using large
![](https://www.researchgate.net/publication/346511372/figure/fig10/AS:963905052483587@1606824281919/The-proposed-semi-supervised-learning-framework-leverages-unlabeled-data-in-two-ways-1_Q320.jpg)
The proposed semi-supervised learning framework leverages
![](https://pub.mdpi-res.com/remotesensing/remotesensing-14-04824/article_deploy/html/images/remotesensing-14-04824-g002.png?1664524956)
Remote Sensing, Free Full-Text
![](https://nlp.fast.ai/images/multifit_bootstrapping.png)
Efficient multi-lingual language model fine-tuning
![](https://miro.medium.com/v2/resize:fit:1400/0*p_vXgNorVabMaiiT.png)
Fine-Tuning LLMs ( Large Language Models )
![](https://miro.medium.com/v2/resize:fit:1400/1*PaFjj_ZCDb1-Uwa3DUfOjA.png)
Supervised Fine-tuning: customizing LLMs
![](https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Ffb9d0144-3952-42db-8382-8e2eb37d917e_1670x640.png)
Understanding and Using Supervised Fine-Tuning (SFT) for Language
![](https://pub.mdpi-res.com/jpm/jpm-13-01703/article_deploy/html/images/jpm-13-01703-ag.png?1703158375)
JPM, Free Full-Text
![](https://arxiv.org/html/2402.13442v1/x5.png)
CoFRIDA: Self-Supervised Fine-Tuning for Human-Robot Co-Painting