Chain-of-thought prompting degrades visual spatial reasoning performance of multimodal LLMs and causes them to hallucinate image details from text alone.
Title resolution pending
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
-
Chain-of-Thought Degrades Visual Spatial Reasoning Capabilities of Multimodal LLMs
Chain-of-thought prompting degrades visual spatial reasoning performance of multimodal LLMs and causes them to hallucinate image details from text alone.