Unverified50% confidenceOpinionExact time
In TI2I, CFG only amplifies on the text condition, which is equivalent to unilaterally strengthening the 'change' direction without any mechanism to protect 'preservation', thereby systematically eroding the constraining force of the image condition.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe conditions of the TI2I task are multimodal: the model simultaneously receives the source image and the editing instruction text, forming an 'alignment-preservation' trade-off, where the text condition drives change and the image condition constrains what stays unchanged.75% similarUnverifiedThe fundamental obstacle to migrating diffusion models to the text domain is that images consist of continuous pixel values while text consists of discrete tokens, making it difficult to directly apply noise to discrete symbols.63% similarUnverifiedCFG scale（无分类器引导强度）控制生成图像对prompt的遵循程度62% similarUnverifiedExisting token merging methods usually rely on text relevance to protect foreground regions, ignoring background objects and local details not mentioned in the prompt, which leads to degraded structural coherence.59% similarUnverified在RAG架构中，文本块比图片具有更高的向量化质量，图片因无法直接向量化成为信息孤岛，预先转化为文字描述能纳入检索体系59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1046698API
curl https://kongchang.com/api/v1/knowledge/claims/1046698MCP
get_claim(id=1046698)