Unverified50% confidenceFactExact time
视觉grounding（视觉定位）指模型将语言描述精准对应到图像具体区域的能力，也被称为Referring Expression Comprehension
1
Sources
50%
Confidence
Long-term
Relevance
8/28/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedVisual primitives replace linguistic descriptions with coordinates, elevating the representation precision of spatial intermediate states from word granularity to pixel granularity.76% similarUnverified视觉幻觉指模型在描述图像时生成与实际视觉内容不符的信息，源于语言模型倾向于用语言先验知识填补视觉信号不明确的空白75% similarUnverified语言模型将风格描述解析为潜在视觉特征向量，图像生成模块在约束条件下进行去噪采样逐步生成图像74% similarUnverified对图像识别来说关键在于特征而非绝对位置，而在语言中顺序决定意义72% similarUnverifiedWhen text instructions precede image input, the Transformer's cross-attention mechanism causes the model to have already formed clear query vectors when encoding the image, preferentially activating visual regions semantically related to the target.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/810992API
curl https://kongchang.com/api/v1/knowledge/claims/810992MCP
get_claim(id=810992)