Unverified50% confidenceFactExact time
In vision-language pretraining, soft labels or label smoothing can be applied to the image-text contrastive loss, replacing hard labels with a mixture of a uniform distribution and the true label, thereby preventing the model from being overconfident about a single incorrect label.
1
Sources
50%
Confidence
Long-term
Relevance
11/7/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWhen using contrastive learning objectives, the damage of noisy labels to vision-language alignment is relatively controllable; generative objectives (such as image captioning) are more sensitive to label errors.78% similarUnverifiedSCARCE's core contribution lies in freeing graph OOD generalization from label dependence, providing a feasible path for graph learning scenarios with scarce labels.76% similarUnverified低质量或过度压缩的训练图像会导致模型在细节生成上产生伪影、色彩失真乃至风格理解偏差70% similarUnverifiedVision-language models are good at identifying "whether there is content" (blank, truncated, garbled), but not good at evaluating "whether the design looks good" (alignment, spacing, visual weight balance).68% similarUnverifiedCLIP 用对比学习（Contrastive Learning）替代了此前的监督分类范式，通过最大化匹配图文对相似度、最小化不匹配对相似度进行自监督学习68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1146921API
curl https://kongchang.com/api/v1/knowledge/claims/1146921MCP
get_claim(id=1146921)