Unverified50% confidenceEventExact time
A new work models visual attention as a continuous distribution centered on the current fixation point that decays with spatial distance, and encodes this distribution into reward model training as an additional supervisory signal.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/11/2026
First Seen
Valid until: 2/9/2027
Sources
Related Entities
Related Claims
UnverifiedAttention visualization shows the distribution of attention weights between tokens; activation probes train linear classifiers on hidden layers to test whether information is decodable, which is correlational evidence; feature attribution measures the contribution of each input dimension to the output.77% similarUnverified扩散模型在训练过程中学习到的是视觉概念的分布式表征,不同IP的视觉元素在潜在空间中以可组合的形式存在77% similarUnverifiedThe core insight of Prompt-to-Prompt is that the attention maps of the cross-attention layers in diffusion models directly control the spatial layout of the image.73% similarUnverifiedInjecting a psychological prior (attention varying gradually with spatial distance) into the reward model as an inductive bias can reduce sample requirements and improve generalization, and is especially suitable for eye-tracking scenarios with limited data scale.73% similarUnverifiedAttention heatmap generation is typically based on predictive models trained on eye-tracking data or computer-vision-based saliency detection algorithms.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1158386API
curl https://kongchang.com/api/v1/knowledge/claims/1158386MCP
get_claim(id=1158386)