TRC² is a brain-inspired decoder-only architecture that localizes fast plasticity and uses thalamic and hippocampal pathways to substantially reduce cumulative forgetting in sequential language model training on streams like C4, WikiText-103, and GSM8K.
Deepseekmoe: Towards ultimate expert specialization in mixture-of-experts language models
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3representative citing papers
LingBot-Video is an open-source MoE video foundation model for embodied intelligence that scales to 120B parameters, integrates robot data, and uses multi-dimensional RL to improve physical plausibility.
Rosetta proposes a composable multimodal pretraining method with MAOP to prevent catastrophic forgetting when expanding modalities beyond standard MoE and MoT approaches.
citing papers explorer
-
Scaling Mixture-of-Experts Video Pretraining for Embodied Intelligence
LingBot-Video is an open-source MoE video foundation model for embodied intelligence that scales to 120B parameters, integrates robot data, and uses multi-dimensional RL to improve physical plausibility.
-
Rosetta: Composable Native Multimodal Pretraining
Rosetta proposes a composable multimodal pretraining method with MAOP to prevent catastrophic forgetting when expanding modalities beyond standard MoE and MoT approaches.