VCE mitigates object hallucination in LVLMs by decomposing activation patterns from contrastive visual inputs via SVD to suppress hallucination subspaces through targeted parameter edits.
Mitigating multilingual hallucination in large vision-language models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
VCE: A zero-cost hallucination mitigation method of LVLMs via visual contrastive editing
VCE mitigates object hallucination in LVLMs by decomposing activation patterns from contrastive visual inputs via SVD to suppress hallucination subspaces through targeted parameter edits.