P\textsuperscript{2}-DPO: Grounding Hallucination in Perceptual Processing via Calibration Direct Preference Optimization

Ruipeng Zhang, Zhihao Li, Haozhang Yuan, C. L. Philip Chen, Tong Zhang

Jun 3, 2026 at 04:00

12 Visningar

0 Kommentarer

arXiv:2606.03376v1 Announce Type: cross Abstract: Hallucination has recently garnered significant research attention in Large Vision-Language Models (LVLMs). Direct Preference Optimization (DPO) aims to learn directly from the corrected preferences provided by humans, thereby addressing the hallucination issue. Despite its success, this paradigm...

Läs hela artikeln hos källan.

Läs originalartikeln

Var detta hjälpsamt?

Dela:

Kommentarer (0)

Vänligen logga in för att publicera en kommentar

Inga kommentarer ännu. Bli först med att kommentera!

Relaterade nyheter

Cryptee Launches End-to-End Encrypted Photo Sharing: Legal Risks, Preventing Abuse, and Their Solution

2 hours ago

Länk kopierad till urklipp

P\textsuperscript{2}-DPO: Grounding Hallucination in Perceptual Processing via Calibration Direct Preference Optimization

Kommentarer (0)

Relaterade nyheter

Cryptee Launches End-to-End Encrypted Photo Sharing: Legal Risks, Preventing Abuse, and Their Solution

The Trouble with Cancer Screening in Healthy Adults

Epics omgjorda launcher blir fem gånger snabbare

[Ekstra] Over én million lærere er flyttet over på en åpen kildekode-plattform

New video game console aims to get kids moving

Bläddra efter kategori