{"total":73,"rows":[{"id":"2412.14219","title":"A Survey on Inference Optimization Techniques for Mixture of Experts Models","created":"2024-12-18","cited_by_count":47,"area":"moe"},{"id":"2504.02263","title":"MegaScale-Infer: Serving Mixture-of-Experts at Scale with Disaggregated Expert Parallelism","created":"2025-04-03","cited_by_count":32,"area":"moe"},{"id":"2503.18292","title":"Jenga: Effective Memory Management for Serving LLM with Heterogeneity","created":"2025-03-24","cited_by_count":25,"area":"inference"},{"id":"2504.15299","title":"D$^{2}$MoE: Dual Routing and Dynamic Scheduling for Efficient On-Device MoE-based LLM Serving","created":"2025-04-17","cited_by_count":20,"area":"moe"},{"id":"2504.08784","title":"SLOs-Serve: Optimized Serving of Multi-SLO LLMs","created":"2025-04-05","cited_by_count":19,"area":null},{"id":"2505.17505","title":"L-MTP: Leap Multi-Token Prediction Beyond Adjacent Context for Large Language Models","created":"2025-05-23","cited_by_count":15,"area":null},{"id":"2506.20675","title":"Utility-Driven Speculative Decoding for Mixture-of-Experts","created":"2025-06-17","cited_by_count":13,"area":"moe"},{"id":"2509.20241","title":"Energy Use of AI Inference, Efficiency Pathways, and Test-Time Scaling","created":"2025-09-24","cited_by_count":13,"area":null},{"id":"2503.06823","title":"eMoE: Task-aware Memory Efficient Mixture-of-Experts-Based (MoE) Model Inference","created":"2025-03-10","cited_by_count":11,"area":"moe"},{"id":"2505.11415","title":"MoE-CAP: Benchmarking Cost, Accuracy and Performance of Sparse Mixture-of-Experts Systems","created":"2025-05-16","cited_by_count":11,"area":"moe"},{"id":"2505.05950","title":"FloE: On-the-Fly MoE Inference on Memory-constrained GPU","created":"2025-05-09","cited_by_count":9,"area":"moe"},{"id":"2602.23036","title":"LLMServingSim 2.0: A Unified Simulator for Heterogeneous and Disaggregated LLM Serving Infrastructure","created":"2026-02-26","cited_by_count":9,"area":"inference"},{"id":"2508.21706","title":"Accelerating Mixture-of-Experts Inference by Hiding Offloading Latency with Speculative Decoding","created":"2025-08-29","cited_by_count":8,"area":"moe"},{"id":"2509.23638","title":"LayerScope: Predictive Cross-Layer Scheduling for Efficient Multi-Batch MoE Inference on Legacy Servers","created":"2025-09-28","cited_by_count":8,"area":"moe"},{"id":"2512.13525","title":"Janus: Disaggregating Attention and Experts for Scalable MoE Inference","created":"2025-12-15","cited_by_count":8,"area":"moe"},{"id":"2510.05497","title":"Patterns behind Chaos: Forecasting Data Movement for Efficient Large-Scale MoE LLM Inference","created":"2025-10-07","cited_by_count":6,"area":"moe"},{"id":"2504.17584","title":"CHIME: A Case for Efficient Long-Context Attention-FC Disaggregated Inference with DIMM-PIM","created":"2025-04-24","cited_by_count":5,"area":"inference"},{"id":"2510.26730","title":"ExpertFlow: Adaptive Expert Scheduling and Memory Coordination for Efficient MoE Inference","created":"2025-10-30","cited_by_count":5,"area":"moe"},{"id":"2511.00868","title":"FlexiCache: Leveraging Temporal Stability of Attention Heads for Efficient KV Cache Management","created":"2025-11-02","cited_by_count":5,"area":"inference"},{"id":"2603.09983","title":"MoE-SpAc: Efficient MoE Inference Based on Speculative Activation Utility in Heterogeneous Edge Scenarios","created":"2026-02-12","cited_by_count":5,"area":"moe"},{"id":"2605.18071","title":"KVDrive: A Holistic Multi-Tier KV Cache Management System for Long-Context LLM Inference","created":"2026-05-18","cited_by_count":5,"area":"inference"},{"id":"2502.09921","title":"A Cost-Effective Near-Storage Processing Solution for Offline Inference of Long-Context LLMs","created":"2025-02-14","cited_by_count":4,"area":"inference"},{"id":"2507.17133","title":"BrownoutServe: SLO-Aware Inference Serving under Bursty Workloads for MoE-based LLMs","created":"2025-07-23","cited_by_count":4,"area":"moe"},{"id":"2601.21198","title":"ZipMoE: Efficient On-Device MoE Serving via Lossless Compression and Cache-Affinity Scheduling","created":"2026-01-29","cited_by_count":4,"area":"moe"},{"id":"2602.00879","title":"Dynamic Expert Sharing: Decoupling Memory from Parallelism in Mixture-of-Experts Diffusion LLMs","created":"2026-01-31","cited_by_count":4,"area":"moe"},{"id":"2603.23049","title":"PCR: A Prefetch-Enhanced Cache Reuse System for Low-Latency RAG Serving","created":"2026-03-24","cited_by_count":4,"area":"rag"},{"id":"2504.03651","title":"Echo: Efficient Co-Scheduling of Hybrid Online-Offline Tasks for Large Language Model Serving","created":"2025-03-01","cited_by_count":3,"area":"inference"},{"id":"2504.09345","title":"MoE-Lens: Towards the Hardware Limit of High-Throughput MoE LLM Serving Under Resource Constraints","created":"2025-04-12","cited_by_count":3,"area":"moe"},{"id":"2508.18983","title":"SMoE: An Algorithm-System Co-Design for Pushing MoE to the Edge via Expert Substitution","created":"2025-08-26","cited_by_count":3,"area":"moe"},{"id":"2509.16857","title":"ShadowServe: Interference-Free KV Cache Fetching for Distributed Prefix Caching","created":"2025-09-21","cited_by_count":3,"area":"inference"},{"id":"2510.02345","title":"Breaking the MoE LLM Trilemma: Dynamic Expert Clustering with Structured Compression","created":"2025-09-27","cited_by_count":3,"area":"moe"},{"id":"2512.17073","title":"Bandwidth-Efficient Adaptive Mixture-of-Experts via Low-Rank Compensation","created":"2025-12-18","cited_by_count":3,"area":"moe"},{"id":"2605.07719","title":"An Efficient Hybrid Sparse Attention with CPU-GPU Parallelism for Long-Context Inference","created":"2026-05-08","cited_by_count":3,"area":"inference"},{"id":"2605.27081","title":"ReMoE: Boosting Expert Reuse through Router Fine-Tuning in Memory-Constrained MoE LLM Inference","created":"2026-05-26","cited_by_count":3,"area":"moe"},{"id":"2608.16157","title":"FreeToken: Efficient Edge-Native MoE Serving with Bandwidth-Adaptive Execution","created":"2026-08-17","cited_by_count":3,"area":"moe"},{"id":"2510.19470","title":"HybridEP: Scaling Expert Parallelism to Cross-Datacenter Scenario via Hybrid Expert/Data Transmission","created":"2025-10-22","cited_by_count":2,"area":"moe"},{"id":"2512.04476","title":"Context-Aware Mixture-of-Experts Inference on CXL-Enabled GPU-NDP Systems","created":"2025-12-04","cited_by_count":2,"area":"moe"},{"id":"2601.20309","title":"SuperInfer: SLO-Aware Rotary Scheduling and Memory Management for LLM Inference on Superchips","created":"2026-01-28","cited_by_count":2,"area":"inference"},{"id":"2602.00328","title":"Harvest: Opportunistic Peer-to-Peer GPU Caching for LLM Inference","created":"2026-01-30","cited_by_count":2,"area":"inference"},{"id":"2602.03495","title":"DALI: A Workload-Aware Offloading Framework for Efficient MoE Inference on Local PCs","created":"2026-02-03","cited_by_count":2,"area":"moe"},{"id":"2603.19172","title":"DyMoE: Dynamic Expert Orchestration with Mixed-Precision Quantization for Efficient MoE Inference on Edge","created":"2026-03-19","cited_by_count":2,"area":"moe"},{"id":"2604.18529","title":"HybridGen: Efficient LLM Generative Inference via CPU-GPU Hybrid Computing","created":"2026-04-20","cited_by_count":2,"area":"inference"},{"id":"2605.02960","title":"MoE-Prefill: Zero Redundancy Overheads in MoE Prefill Serving","created":"2026-05-03","cited_by_count":2,"area":"moe"},{"id":"2606.21868","title":"WiSP: A Working-Set View of Mixture-of-Experts Serving on Extremely Low-Resource Hardware","created":"2026-06-20","cited_by_count":2,"area":"moe"},{"id":"2508.19078","title":"Federated Fine-Tuning of Sparsely-Activated Large Language Models on Resource-Constrained Devices","created":"2025-08-26","cited_by_count":1,"area":"moe"},{"id":"2512.03927","title":"OD-MoE: On-Demand Expert Loading for Cacheless Edge-Distributed MoE Inference","created":"2025-12-03","cited_by_count":1,"area":"moe"},{"id":"2512.21487","title":"Efficient MoE Inference with Fine-Grained Scheduling of Disaggregated Expert Parallelism","created":"2025-12-25","cited_by_count":1,"area":"moe"},{"id":"2602.09725","title":"Efficient Remote KV Cache Reuse with GPU-native Video Codec","created":"2026-02-10","cited_by_count":1,"area":"inference"},{"id":"2606.18600","title":"ShuntServe: Cost-Efficient LLM Serving on Heterogeneous Spot GPU Clusters","created":"2026-06-17","cited_by_count":1,"area":"inference"},{"id":"2606.29982","title":"Beyond Uniform Experts: Cost-Aware Expert Execution for Efficient Multi-Device MoE Inference","created":"2026-06-29","cited_by_count":1,"area":"moe"},{"id":"2412.07067","title":"MoE-CAP: Benchmarking Cost, Accuracy and Performance of Sparse Mixture-of-Experts Systems","created":"2024-12-10","cited_by_count":0,"area":"moe"},{"id":"2506.03296","title":"APEX: Asynchronous Parallel CPU-GPU Execution for Online LLM Inference on Constrained GPUs","created":"2025-06-03","cited_by_count":0,"area":"inference"},{"id":"2601.03992","title":"A Scheduling Framework for Efficient MoE Inference on Edge GPU-NDP Systems","created":"2026-01-07","cited_by_count":0,"area":"moe"},{"id":"2602.08818","title":"FlexMoRE: A Flexible Mixture of Rank-heterogeneous Experts for Efficient Federatedly-trained Large Language Models","created":"2026-02-09","cited_by_count":0,"area":"moe"},{"id":"2603.27624","title":"Expert Streaming: Accelerating Low-Batch MoE Inference via Multi-chiplet Architecture and Dynamic Expert Trajectory Scheduling","created":"2026-03-29","cited_by_count":0,"area":"moe"},{"id":"2604.02715","title":"FluxMoE: Decoupling Expert Residency for High-Performance MoE Serving","created":"2026-04-03","cited_by_count":0,"area":"moe"},{"id":"2604.14626","title":"ELMoE-3D: Leveraging Intrinsic Elasticity of MoE for Hybrid-Bonding-Enabled Self-Speculative Decoding in On-Premises Serving","created":"2026-04-16","cited_by_count":0,"area":"moe"},{"id":"2604.26334","title":"Efficient, VRAM-Constrained xLM Inference on Clients","created":"2026-04-29","cited_by_count":0,"area":"mllm"},{"id":"2605.02444","title":"M\\textsuperscript{4}Fuse: Lightweight State-Space MoE with a Cross-Scale Gating Bridge for Brain Tumor Segmentation","created":"2026-05-04","cited_by_count":0,"area":"moe"},{"id":"2605.11335","title":"ChunkFlow: Communication-Aware Chunked Prefetching for Layerwise Offloading in Distributed Diffusion Transformer Inference","created":"2026-05-11","cited_by_count":0,"area":"diffusion"},{"id":"2605.16637","title":"HexAGenT: Efficient Agentic LLM Serving via Workflow- and Heterogeneity-Aware Scheduling","created":"2026-05-15","cited_by_count":0,"area":"inference"},{"id":"2605.17889","title":"CoX-MoE: Coalesced Expert Execution for High-Throughput MoE Inference with AMX-Enabled CPU-GPU Co-Execution","created":"2026-05-18","cited_by_count":0,"area":"moe"},{"id":"2606.21712","title":"BatchGen: An Architecture for Scalable and Efficient Batch Inference","created":"2026-06-19","cited_by_count":0,"area":null},{"id":"2607.10183","title":"Automated Tensor Scheduling for Hybrid CPU-GPU LLM Inference on Consumer Devices","created":"2026-07-11","cited_by_count":0,"area":null},{"id":"2607.23099","title":"Decoding the Skew: Distribution-Aware MoE Inference with Adaptive Kernel Dispatch","created":"2026-07-25","cited_by_count":0,"area":"moe"},{"id":"2608.13962","title":"MoE Expert Execution in Disaggregated LLM Serving with a High-Bandwidth ReRAM Near-Memory Architecture","created":"2026-08-14","cited_by_count":0,"area":"moe"},{"id":"2608.15383","title":"Every Expert Counts: ExactMoE for Memory-Efficient W4A16 Inference","created":"2026-08-15","cited_by_count":0,"area":"moe"},{"id":"2609.06551","title":"EStream: Fast and Memory-Efficient MoE Prefill through Expert Virtualization on Mobile NPUs","created":"2026-09-06","cited_by_count":0,"area":"moe"},{"id":"2609.13537","title":"Dynamic HBM Repartitioning for Multi-Turn MoE Serving","created":"2026-09-11","cited_by_count":0,"area":"moe"},{"id":"2609.13592","title":"BOOST: Concurrent Access to Host Memory and HBM to Accelerate LLM Inference","created":"2026-09-11","cited_by_count":0,"area":null},{"id":"2609.14507","title":"Physically Partitioned KVCache Format for CPU--GPU Load Balancing in MoE Inference","created":"2026-09-13","cited_by_count":0,"area":"moe"},{"id":"2609.18110","title":"SSD-LLaMA: SSD-Native Inference for Trillion-Parameter MoE at 1+ Token/s on a Consumer PC","created":"2026-09-16","cited_by_count":0,"area":"moe"},{"id":"2609.34380","title":"DPS: Dual-Mode Precision LLM Serving with Semi-Unified Memory","created":"2026-09-28","cited_by_count":0,"area":"inference"}],"next_cursor":null}