VGA constructs precise visual grounding from token semantics to guide MLLM attention toward relevant regions, dynamically suppressing described areas in captioning, and achieves SOTA dehallucination with negligible overhead.
Cracking the code of hallucination in lvlms with vision-aware head divergence
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Tell Model Where to Look: Mitigating Hallucinations in MLLMs by Vision-Guided Attention
VGA constructs precise visual grounding from token semantics to guide MLLM attention toward relevant regions, dynamically suppressing described areas in captioning, and achieves SOTA dehallucination with negligible overhead.