home/categories/llm-ai/davila7-claude-code-templates-cli-tool-components-skills-ai-research-post-training-simpo-skill-md
llm-aidata-ai
simpo-training
Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.
maintainer
davila7
更新於 1/20/2026
星標
17577
分支
1576
quick start
Installation and usage
Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.
安裝
$ install --globalskills.sh
使用
安裝後，您可以通過在終端運行以下命令來使用此技能：
skills use simpo-training