Termsslug: rl-fine-tuning

RL Fine-tuning

强化学习微调

Adjusting pre-trained model parameters using reward signals to optimize specific goals

Curated Intelligence

2 items · Page 1/1
Hacker News AIT2·Industry
45 pts
Anthropic Blocks Users for Model Training Activities

Anthropic has implemented account bans and downgrades for users engaging in model training activities like RFT and SFT, even when using APIs solely for automation rather than direct distillation. This move has sparked backlash from developers who argue it hinders research automation and highlights perceived hypocrisy given Anthropic's own reliance on public data.

Hugging Face BlogT1·Tools
25 pts
TRL Enables DDPO Fine-tuning for Stable Diffusion

The Hugging Face TRL library now supports DDPO (Denoising Diffusion Policy Optimization), enabling reinforcement learning-based fine-tuning of Stable Diffusion models. This feature aims to align generated images with human preferences and is currently in the tool integration phase.