cs.AI, cs.CL, cs.LG

Toward Consistent World Models with Multi-Token Prediction and Latent Semantic Enhancement

arXiv:2604.06155v1 Announce Type: cross
Abstract: Whether Large Language Models (LLMs) develop coherent internal world models remains a core debate. While conventional Next-Token Prediction (NTP) focuses on one-step-ahead supervision, Multi-Token Pred…