Pith. sign in

Title resolution pending

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.CV 1

years

2025 1

verdicts

CONDITIONAL 1

representative citing papers

Improving Alignment in LVLMs with Debiased Self-Judgment

cs.CV · 2025-08-28 · conditional · novelty 6.0

A contrastive self-judgment score that subtracts a model's image-free confidence from its visual confidence is used to guide decoding, safety moderation, and DPO training, improving hallucination and safety metrics across LVLMs.

citing papers explorer

Showing 1 of 1 citing paper.

  • Improving Alignment in LVLMs with Debiased Self-Judgment cs.CV · 2025-08-28 · conditional · none · ref 2

    A contrastive self-judgment score that subtracts a model's image-free confidence from its visual confidence is used to guide decoding, safety moderation, and DPO training, improving hallucination and safety metrics across LVLMs.