Breaking the 100M Token Limit: EverMind's MSA Architecture Achieves Efficient End-to-End Long-Term Memory for LLMs

The research introduces a novel memory architecture called MSA (Memory Sparse Attention). Through a combination of the Memory Sparse Attention mechanism, Document-wise RoPE for extreme context ...
IEEE

Multi-Agent Cooperative Exploration via Weighted Conditional Mutual Information in Sparse Reward Tasks

Abstract: Multi-agent Reinforcement Learning (MARL) algorithms have shown remarkable success in tasks requiring cooperation. However, in sparse reward scenarios, traditional MARL algorithms struggle ...
IEEE

Multi-Vector Model Predictive Control for NPC Inverter-Fed PMSM Drives Based on Fast Vector Selection

Abstract: Model predictive control (MPC) has recently been considered for permanent magnet synchronous motor (PMSM) drive applications. However, the traditional finite-set MPC suffers from such ...