psp-dada/Uni-DPO download history
psp-dada/Uni-DPO is an image text to text dataset on the Hugging Face Hub. In the last 30 days it was downloaded 85 times (12 in the last 7 days), and 896 times in total. It ranks #127,810 among datasets by monthly downloads.
Dataset Card for ICLR 2026 | Uni-DPO: A Unified Paradigm for Dynamic Preference Optimization of LLMs 中文 | English Paper Abstract Direct Preference Optimization (DPO) has emerged as a cornerstone of reinforcement learning from human feedback (RLHF) due to its si
Models trained on Uni-DPO
9 models list it as training data.
- psp-dada/Qwen2.5-Math-7B-Uni-DPO 19 downloads in 30 days
- psp-dada/Llama-3-8B-Base-SFT-Uni-DPO-v2-GPT-4 18 downloads in 30 days
- psp-dada/Llama-3-8B-Instruct-Uni-DPO-v2-ArmoRM 17 downloads in 30 days
- psp-dada/Llama-3-8B-Base-SFT-Uni-DPO 16 downloads in 30 days
- psp-dada/Llama-3-8B-Instruct-Uni-DPO 14 downloads in 30 days
- psp-dada/Gemma2-9B-IT-Uni-DPO 13 downloads in 30 days
- psp-dada/Llama-3-8B-Instruct-Uni-DPO-v2-GPT-4o 12 downloads in 30 days
- psp-dada/Qwen2.5-7B-Uni-DPO 11 downloads in 30 days
- psp-dada/Llama-3-8B-Base-SFT-Uni-DPO-v2-Qwen 7 downloads in 30 days
Open psp-dada/Uni-DPO on Hugging Face
Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.