SPPO - a UCLA-AGI Collection

UCLA-AGI 's Collections

zephyr-7b-sft-full-SPIN

SPPO

SPPO

updated Jun 29

Self-Play Preference Optimization

UCLA-AGI/Mistral7B-PairRM-SPPO

Text Generation • Updated May 7 • 3.49k • 6
UCLA-AGI/Mistral7B-PairRM-SPPO-Iter1

Text Generation • Updated May 6 • 2.65k • 2
UCLA-AGI/Mistral7B-PairRM-SPPO-Iter2

Text Generation • Updated May 6 • 7.93k • 1
UCLA-AGI/Mistral7B-PairRM-SPPO-Iter3

Text Generation • Updated May 7 • 7.94k • 5
UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter1

Text Generation • Updated Jun 25 • 4.84k • 1
UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter2

Text Generation • Updated Jun 25 • 7.92k
UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3

Text Generation • Updated Jun 28 • 7.63k • 77
UCLA-AGI/Gemma-2-9B-It-SPPO-Iter3

Text Generation • Updated Jul 1 • 9.09k • 118
UCLA-AGI/Gemma-2-9B-It-SPPO-Iter2

Text Generation • Updated Jul 1 • 6.56k • 2
UCLA-AGI/Gemma-2-9B-It-SPPO-Iter1

Text Generation • Updated Jul 1 • 5.16k • 3