CHAIR-DPO - a aimagelab Collection

aimagelab 's Collections

DICE

ReT-2

RAID

ReT

CHAIR-DPO

updated 7 days ago

[BMVC 2025] Mitigating Hallucinations in Multimodal LLMs via Object-aware Preference Optimization - Models and Datasets

Mitigating Hallucinations in Multimodal LLMs via Object-aware Preference Optimization

Paper • 2508.20181 • Published Aug 27
aimagelab/CHAIR-DPO_LLaVA-MORE-8B

8B • Updated 8 days ago • 4
aimagelab/CHAIR-DPO_LLaVA-1.5-7B

7B • Updated 8 days ago • 4
aimagelab/CHAIR-DPO_preference_datasets

Viewer • Updated about 13 hours ago • 1.39M • 20