{"total":49,"rows":[{"id":"2502.05172","title":"Joint MoE Scaling Laws: Mixture of Experts Can Be Memory Efficient","created":"2025-02-07","cited_by_count":31,"area":"moe"},{"id":"2410.19034","title":"Mixture of Parrots: Experts improve memorization more than reasoning","created":"2024-10-24","cited_by_count":21,"area":"moe"},{"id":"2508.02317","title":"VeOmni: Scaling Any Modality Model Training with Model-Centric Distributed Recipe Zoo","created":"2025-08-04","cited_by_count":20,"area":"moe"},{"id":"2605.09165","title":"Sparse Layers are Critical to Scaling Looped Language Models","created":"2026-05-09","cited_by_count":11,"area":"moe"},{"id":"2506.21103","title":"Learning to Skip the Middle Layers of Transformers","created":"2025-06-26","cited_by_count":10,"area":"moe"},{"id":"2510.17483","title":"ReXMoE: Reusing Experts with Minimal Overhead in Mixture-of-Experts","created":"2025-10-20","cited_by_count":10,"area":"moe"},{"id":"2509.26520","title":"Training Matryoshka Mixture-of-Experts for Elastic Inference-Time Expert Utilization","created":"2025-09-30","cited_by_count":7,"area":"moe"},{"id":"2511.02237","title":"Opportunistic Expert Activation: Batch-Aware Expert Routing for Faster Decode Without Retraining","created":"2025-11-04","cited_by_count":7,"area":"moe"},{"id":"2505.07260","title":"UMoE: Unifying Attention and FFN with Shared Experts","created":"2025-05-12","cited_by_count":6,"area":"moe"},{"id":"2509.25913","title":"Understanding the Mixture-of-Experts with Nadaraya-Watson Kernel","created":"2025-09-30","cited_by_count":6,"area":"moe"},{"id":"2503.05029","title":"Continual Pre-training of MoEs: How robust is your router?","created":"2025-03-06","cited_by_count":5,"area":"moe"},{"id":"2506.05767","title":"dots.llm1 Technical Report","created":"2025-06-06","cited_by_count":5,"area":"moe"},{"id":"2602.01203","title":"Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse","created":"2026-02-01","cited_by_count":5,"area":"moe"},{"id":"2410.11222","title":"Quadratic Gating Mixture of Experts: Statistical Insights into Self-Attention","created":"2024-10-15","cited_by_count":4,"area":"moe"},{"id":"2507.09205","title":"From Curated Data to Scalable Models: Continual Pre-training of Dense and MoE Large Language Models for Tibetan","created":"2025-07-12","cited_by_count":4,"area":"moe"},{"id":"2507.18671","title":"Innovator: Scientific Continued Pretraining with Fine-grained MoE Upcycling","created":"2025-07-24","cited_by_count":4,"area":"moe"},{"id":"2603.18297","title":"Path-Constrained Mixture-of-Experts","created":"2026-03-18","cited_by_count":4,"area":"moe"},{"id":"2603.11535","title":"Expert Threshold Routing for Autoregressive Language Modeling with Dynamic Computation Allocation and Load Balancing","created":"2026-03-12","cited_by_count":3,"area":"moe"},{"id":"2604.00421","title":"Self-Routing: Parameter-Free Expert Routing from Hidden States","created":"2026-04-01","cited_by_count":3,"area":"moe"},{"id":"2604.25578","title":"Marco-MoE: Open Multilingual Mixture-of-Expert Language Models with Efficient Upcycling","created":"2026-04-28","cited_by_count":3,"area":"moe"},{"id":"2606.25971","title":"Improving Neural Network Training by Decoupling the Magnitude and Direction of Weight Vectors","created":"2026-06-24","cited_by_count":3,"area":"moe"},{"id":"2504.15527","title":"Compass-V2 Technical Report","created":"2025-04-22","cited_by_count":2,"area":"moe"},{"id":"2512.16248","title":"Sigma-MoE-Tiny Technical Report","created":"2025-12-18","cited_by_count":2,"area":"moe"},{"id":"2502.12455","title":"DSMoE: Matrix-Partitioned Experts with Dynamic Routing for Computation-Efficient Dense LLMs","created":"2025-02-18","cited_by_count":1,"area":"moe"},{"id":"2510.08008","title":"Beyond Sunk Costs: Boosting LLM Pre-training Efficiency via Orthogonal Growth of Mixture-of-Experts","created":"2025-10-09","cited_by_count":1,"area":"moe"},{"id":"2601.17680","title":"$\\infty$-MoE: Generalizing Mixture of Experts to Infinite Experts","created":"2026-01-25","cited_by_count":1,"area":"moe"},{"id":"2605.11255","title":"HEBATRON: A Hebrew-Specialized Open-Weight Mixture-of-Experts Language Model","created":"2026-05-11","cited_by_count":1,"area":"moe"},{"id":"2605.28207","title":"Pruning and Distilling Mixture-of-Experts into Dense Language Models","created":"2026-05-27","cited_by_count":1,"area":"moe"},{"id":"2608.06398","title":"EntropyMoE: Entropy-Aware Sparse Expert Routing for Tokenizer-Free LLMs","created":"2026-07-31","cited_by_count":1,"area":"moe"},{"id":"2609.35751","title":"How to Loop MoE: Flatten the Experts, Untie the Attention","created":"2026-09-28","cited_by_count":1,"area":"moe"},{"id":"2601.08215","title":"Towards Principled Design of Mixture-of-Experts Language Models under Memory and Inference Constraints","created":"2026-01-13","cited_by_count":0,"area":"moe"},{"id":"2602.11543","title":"Pretraining A Large Language Model using Distributed GPUs: A Memory-Efficient Decentralized Paradigm","created":"2026-02-12","cited_by_count":0,"area":"moe"},{"id":"2605.09516","title":"Mixture of Layers with Hybrid Attention","created":"2026-05-10","cited_by_count":0,"area":"moe"},{"id":"2605.10933","title":"DECO: Sparse Mixture-of-Experts with Dense-Comparable Performance on End-Side Devices","created":"2026-05-11","cited_by_count":0,"area":"moe"},{"id":"2605.25698","title":"How Should LLMs Consume High-Quality Data? Optimal Data Scheduling via Quality-Aware Functional Scaling Laws","created":"2026-05-25","cited_by_count":0,"area":"moe"},{"id":"2605.31268","title":"Mellum2 Technical Report","created":"2026-05-29","cited_by_count":0,"area":"moe"},{"id":"2607.22577","title":"cMoLLM at Scale: Horizontal Scaling Laws for Mixture-of-LLMs","created":"2026-06-06","cited_by_count":0,"area":"moe"},{"id":"2608.15102","title":"A Declarative-Procedural Perspective on Expert Routing in Bilingual Mixture-of-Experts Language Models","created":"2026-08-15","cited_by_count":0,"area":"moe"},{"id":"2608.30567","title":"TuringLLM: Efficiently Scaling Foundation Models Toward Physical AI","created":"2026-08-31","cited_by_count":0,"area":"moe"},{"id":"2609.16179","title":"Z-Loss Backward Geometry in Dense Output Heads and Sparse Routers","created":"2026-07-31","cited_by_count":0,"area":"moe"},{"id":"2609.29618","title":"An Exploratory Ablation of a Small MLA--SSM Hybrid Language Model","created":"2026-08-28","cited_by_count":0,"area":"moe"},{"id":"2609.34634","title":"A Persistent State for Auditable Mixture-of-Experts Routing","created":"2026-09-28","cited_by_count":0,"area":"moe"},{"id":"2609.35481","title":"TopoEP: Topology-Aware Load Balancing for Expert-Parallel MoE Training","created":"2026-09-28","cited_by_count":0,"area":"moe"},{"id":"2610.01153","title":"Looping Beyond Twice: A Scalable Recipe for Looped Mixture-of-Experts","created":"2026-10-01","cited_by_count":0,"area":"moe"},{"id":"2610.02584","title":"Dense Mixture-of-Experts as a Reparameterized Wide FFN: A Granularity Sweep at Fixed Compute","created":"2026-10-01","cited_by_count":0,"area":"moe"},{"id":"2610.11575","title":"Smoothing the Top-k Exposure Boundary for Sparse Mixture-of-Experts","created":"2026-10-08","cited_by_count":0,"area":"moe"},{"id":"2610.17429","title":"Monarch-Muon: Structure-Aware Muon for Fast and Memory-Efficient Pretraining","created":"2026-10-12","cited_by_count":0,"area":"moe"},{"id":"2610.19975","title":"LatentLoop: Scaling Looped MoE Models with Asymmetric Compute Allocation","created":"2026-10-14","cited_by_count":0,"area":"moe"},{"id":"2610.20247","title":"Scaling Multilingual Mixture-of-Experts: How Language Coverage Changes Capacity and Compute Allocation","created":"2026-10-14","cited_by_count":0,"area":"moe"}],"next_cursor":null}