Peng Shangpin's picture

Peng Shangpin

psp-dada

·

https://github.com/pspdada

AI & ML interests

Multimodal Large Language Models, Preference Optimization, Algorithm

Recent Activity

upvoted a paper 2 days ago

Uni-OPD: Unifying On-Policy Distillation with a Dual-Perspective Recipe

authored a paper 4 days ago

Uni-OPD: Unifying On-Policy Distillation with a Dual-Perspective Recipe

updated a model 3 months ago

psp-dada/Qwen2.5-Math-7B-Uni-DPO

View all activity

Organizations

None yet

authored a paper 4 days ago

Uni-OPD: Unifying On-Policy Distillation with a Dual-Perspective Recipe

Paper • 2605.03677 • Published 13 days ago • 5

authored a paper 6 months ago

HunyuanOCR Technical Report

Paper • 2511.19575 • Published Nov 24, 2025 • 22

authored 2 papers 10 months ago

Mitigating Object Hallucinations via Sentence-Level Early Intervention

Paper • 2507.12455 • Published Jul 16, 2025 • 9

Omni-DPO: A Dual-Perspective Paradigm for Dynamic Preference Learning of LLMs

Paper • 2506.10054 • Published Jun 11, 2025 • 3