Unverified50% confidenceConceptExact time
In interpretability research, visual attribution refers to methods that attribute model outputs to specific image regions or visual features, with common approaches including gradient saliency, attention rollout, and perturbation contrast.
1
Sources
50%
Confidence
Long-term
Relevance
10/20/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedFor vision-language models, image resolution affects the recognition accuracy of chart and document details, and the number of visual tokens directly affects the remaining usable context window.77% similarUnverifiedResponsive Visualization is a research direction with existing academic accumulation, referring to charts that dynamically adjust their layout and content density based on screen size and device type.77% similarUnverified将稀疏可解释性方法拓展到视觉领域面临独特挑战，因为视觉特征具有更强的空间相关性、层次性和组合性76% similarUnverified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 Comprehension76% similarUnverifiedMulti-reference image input can decouple different visual dimensions, giving the model more independent constraint signals when processing character identity and scene environment, and reducing mutual interference between visual elements.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1073350API
curl https://kongchang.com/api/v1/knowledge/claims/1073350MCP
get_claim(id=1073350)