Skip to content
QbitAI · WeChat

NUS and collaborators propose ViF to curb visual hallucination snowballing in multi-agent systems

终结多智能体视觉幻觉“滚雪球”!新国立等提出ViF:无需改造模型,即插即用

NUS LV-Lab and collaborators proposed ViF, accepted to ICLR 2026. Across 8 benchmarks, 4 MAS structures, and 10 VLMs, it reports 2.4%–3.8% average gains. ViF replaces text-only passing with visual relay tokens and layered attention redistribution, cutting HS by over 30% on average and nearly 40% in ring topology.

Why it matters: NUS and other teams propose ViF, packing visual information into relay tokens inside multi-agent dialogue so models need no changes to curb hallucination buildup from text-only handoffs.

Read the original ↗Export Markdown