LTS-FS locates hallucination-relevant layers in LVLMs via causal attribution on a constructed dataset and applies sparse layerwise feature steering to mitigate hallucinations while preserving general task performance.
The mask threshold rs is selected to be 0.5, as shown in Tab.5 of main text
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
-
Locate-then-Sparsify: Attribution Guided Sparse Strategy for Visual Hallucination Mitigation
LTS-FS locates hallucination-relevant layers in LVLMs via causal attribution on a constructed dataset and applies sparse layerwise feature steering to mitigate hallucinations while preserving general task performance.