MPD reduces hallucinations in LVLMs by 23.4% while retaining 97.4% of general capability through semantic disentanglement and selective parameter updates.
ICLR 2023 Workshop on Multimodal Representation Learning: Perks and Pitfalls , year=
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
-
Mitigating Hallucinations in Large Vision-Language Models without Performance Degradation
MPD reduces hallucinations in LVLMs by 23.4% while retaining 97.4% of general capability through semantic disentanglement and selective parameter updates.