Unverified50% confidenceFactExact time
JEPA (Joint-Embedding Predictive Architecture) is the self-supervised learning framework championed by Yann LeCun's team, which predicts the representation of another part of the input in latent space rather than predicting raw content.
1
Sources
50%
Confidence
Long-term
Relevance
10/3/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedLeCun在2022年提出的JEPA（Joint Embedding Predictive Architecture）在表征空间中直接预测目标嵌入，而非重构原始输入77% similarUnverifiedJEPA (Joint Embedding Predictive Architecture), which Yann LeCun has strongly advocated for in recent years, is an extension of the world model approach.76% similarUnverifiedThe JEPA framework proposed by Yann LeCun centers on predictive perception and attempts to replace the traditional generative or discriminative paradigm.74% similarUnverifiedJEPA 与自监督学习谱系的关键区别在于明确将'预测'而非'重建'或'对比'作为核心训练信号，且预测目标是经过编码的抽象表示71% similarUnverifiedMol-JEPA is a multimodal JEPA (Joint Embedding Predictive Architecture) foundation model for the molecular domain.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/969694API
curl https://kongchang.com/api/v1/knowledge/claims/969694MCP
get_claim(id=969694)