Learning from Reference Answers: Versatile Language Model Alignment without Binary Human Preference Data
Paper
•
2504.09895
•
Published
•
1
RefAlign: RL with Similarity-based Rewards
Datasets and models in: Learning from Reference Answers: Versatile Language Model Alignment without Binary Human Preference Data.
-
-
mzhaoshuai/Mistral-7B-v0.1-conf-sft
Text Generation • Updated • 55 -
mzhaoshuai/Llama-3.3-70B-Inst-awq_ultrafeedback_1in3
Viewer • Updated • 61.1k • 48 -
mzhaoshuai/Llama-3.3-70B-Inst-awq_SafeRLHF
Preview • Updated • 29 -
mzhaoshuai/zephyr-7b-alpha-conf-sft
Text Generation • Updated • 39 -
mzhaoshuai/alpaca-7b-ref-bertscore
Text Generation • 7B • Updated • 66 -
mzhaoshuai/Llama-2-7b-hf-conf-sft
Text Generation • Updated • 69 -
mzhaoshuai/zephyr-7b-alpha-conf-refalign
Updated • 46 -
mzhaoshuai/Mistral-7B-v0.1-conf-refalign
Text Generation • Updated • 46 -
mzhaoshuai/alpaca-7b-ref-meteor
Text Generation • 7B • Updated • 46 -
mzhaoshuai/Llama-2-13b-hf-conf-sft
Text Generation • Updated • 51 -
mzhaoshuai/Llama-2-13b-hf-conf-refalign
Updated • 28 -
mzhaoshuai/Llama-2-7b-hf-conf-refalign
Text Generation • Updated • 43 -
mzhaoshuai/Mistral-7B-Instruct-v0.2-ref-simpo
Text Generation • 7B • Updated • 35 -
mzhaoshuai/Mistral-7B-Instruct-v0.2-refalign
Text Generation • 7B • Updated • 42 -
mzhaoshuai/NQ-Subset-500
Viewer • Updated • 500 • 25 -
mzhaoshuai/Llama-3-8B-Instruct-ref-simpo
Text Generation • 8B • Updated • 43 -
mzhaoshuai/Llama-3-8B-Instruct-refalign
Text Generation • 8B • Updated • 37 -
mzhaoshuai/llama3-ultrafeedback-bertscore-bart-large-mnli
Viewer • Updated • 60.9k • 34