Tag: Fine-tuning

1. Teaching a model to refuse the harmful half of a topic, not the whole thing (huggingface.co)
   AI · just now · September 14, 2026
2. A 350M model learns reliable JSON with 100 steps of reinforcement learning (huggingface.co)
   AI · 1 day ago · September 12, 2026
3. A specialized OCR model still beats newer general ones (huggingface.co)
   Engineering · 1 month ago · July 27, 2026
4. One transformer layer can match full RL fine-tuning (arxiv.org)
   AI · 2 months ago · July 3, 2026
5. LoRA is the default fine-tuning method, but maybe not the best one (huggingface.co)
   AI · 2 months ago · June 19, 2026

← all tags