New papers tackle hallucination in vision-language models

Multiple arXiv papers propose methods to reduce hallucination in LVLMs, including Dynamic Alignment Compensation, PatchGate, and attention-head targeting in LLaVA. SHROOM-Visions 2026 shared task overview also released.