{"total":57,"rows":[{"id":"2507.19427","title":"Step-3 is Large yet Affordable: Model-system Co-design for Cost-effective Decoding","created":"2025-07-25","cited_by_count":55,"area":"moe"},{"id":"2505.12335","title":"Is Artificial Intelligence Generated Image Detection a Solved Problem?","created":"2025-05-18","cited_by_count":51,"area":"diffusion"},{"id":"2502.19811","title":"Comet: Fine-grained Computation-communication Overlapping for Mixture-of-Experts","created":"2025-02-27","cited_by_count":49,"area":"moe"},{"id":"2410.18038","title":"POD-Attention: Unlocking Full Prefill-Decode Overlap for Faster LLM Inference","created":"2024-10-23","cited_by_count":47,"area":null},{"id":"2409.13221","title":"Optimizing RLHF Training for Large Language Models with Stage Fusion","created":"2024-09-20","cited_by_count":43,"area":"alignment"},{"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":"2505.11432","title":"MegaScale-MoE: Large-Scale Communication-Efficient Training of Mixture-of-Experts Models in Production","created":"2025-05-16","cited_by_count":31,"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":"2603.07685","title":"Scalable Training of Mixture-of-Experts Models with Megatron Core","created":"2026-03-08","cited_by_count":19,"area":"moe"},{"id":"2501.10714","title":"FSMoE: A Flexible and Scalable Training System for Sparse Mixture-of-Experts Models","created":"2025-01-18","cited_by_count":17,"area":"moe"},{"id":"2509.16293","title":"Robust LLM Training Infrastructure at ByteDance","created":"2025-09-19","cited_by_count":16,"area":null},{"id":"2503.06433","title":"Seesaw: High-throughput LLM Inference via Model Re-sharding","created":"2025-03-09","cited_by_count":15,"area":"inference"},{"id":"2505.11329","title":"TokenWeave: Efficient Compute-Communication Overlap for Distributed LLM Inference","created":"2025-05-16","cited_by_count":14,"area":null},{"id":"2504.19519","title":"Efficient and Adaptable Overlapping for Computation and Communication via Signaling and Reordering","created":"2025-04-28","cited_by_count":10,"area":null},{"id":"2409.15241","title":"Domino: Eliminating Communication in LLM Training via Generic Tensor Slicing and Overlapping","created":"2024-09-23","cited_by_count":9,"area":null},{"id":"2501.06589","title":"Ladder-residual: parallelism-aware architecture for accelerating large model inference with communication overlapping","created":"2025-01-11","cited_by_count":9,"area":null},{"id":"2501.08192","title":"PRESERVE: Prefetching Model Weights and KV-Cache in Distributed LLM Serving","created":"2025-01-14","cited_by_count":9,"area":"inference"},{"id":"2512.03870","title":"Reconstructing KV Caches with Cross-layer Fusion For Enhanced Transformers","created":"2025-12-03","cited_by_count":6,"area":"inference"},{"id":"2604.19241","title":"UniEP: Unified Expert-Parallel MoE MegaKernel for LLM Training","created":"2026-04-21","cited_by_count":5,"area":"moe"},{"id":"2510.20171","title":"Collective Communication for 100k+ GPUs","created":"2025-10-23","cited_by_count":4,"area":null},{"id":"2409.11155","title":"ISO: Overlap of Computation and Communication within Seqenence For LLM Inference","created":"2024-09-04","cited_by_count":3,"area":null},{"id":"2503.04983","title":"Leveraging Large Language Models For Scalable Vector Graphics Processing: A Review","created":"2025-03-06","cited_by_count":3,"area":null},{"id":"2505.15765","title":"Constructing a 3D Scene from a Single Image","created":"2025-05-21","cited_by_count":3,"area":null},{"id":"2512.08905","title":"Self-Evolving 3D Scene Generation from a Single Image","created":"2025-12-09","cited_by_count":3,"area":null},{"id":"2601.17654","title":"Kareus: Joint Reduction of Dynamic and Static Energy in Large Model Training","created":"2026-01-25","cited_by_count":3,"area":null},{"id":"2605.08639","title":"ReLibra: Routing-Replay-Guided Load Balancing for MoE Training in Reinforcement Learning","created":"2026-05-09","cited_by_count":3,"area":"moe"},{"id":"2504.09590","title":"Efficient LLM Serving on Hybrid Real-time and Best-effort Requests","created":"2025-04-13","cited_by_count":2,"area":"inference"},{"id":"2601.20273","title":"SwiftFusion: Scalable Sequence Parallelism for Distributed Inference of Diffusion Transformers on GPUs","created":"2026-01-28","cited_by_count":2,"area":"diffusion"},{"id":"2603.02376","title":"CUCo: An Agentic Framework for Compute and Communication Co-design","created":"2026-03-02","cited_by_count":2,"area":"agents"},{"id":"2604.17172","title":"UCCL-Zip: Lossless Compression Supercharged GPU Communication","created":"2026-04-19","cited_by_count":2,"area":null},{"id":"2504.04279","title":"Could AI Trace and Explain the Origins of AI-Generated Images and Text?","created":"2025-04-05","cited_by_count":1,"area":null},{"id":"2505.19196","title":"CoCA: Step-level Reward for Free in RL-based T2I Diffusion Model Fine-tuning","created":"2025-05-25","cited_by_count":1,"area":"diffusion"},{"id":"2512.10236","title":"Design Space Exploration of DMA based Finer-Grain Compute Communication Overlap","created":"2025-12-11","cited_by_count":1,"area":null},{"id":"2601.20461","title":"Exploiting the Final Component of Generator Architectures for AI-Generated Image Detection","created":"2026-01-28","cited_by_count":1,"area":null},{"id":"2602.07306","title":"Parallel Track Transformers: Enabling Fast GPU Inference with Reduced Synchronization","created":"2026-02-07","cited_by_count":1,"area":null},{"id":"2605.17879","title":"Guard: Scalable Straggler Detection and Node Health Management for Large-Scale Training","created":"2026-05-18","cited_by_count":1,"area":null},{"id":"2607.18002","title":"ExpertPlex: A High-Goodput Disaggregated Serving System for MoE LLMs with Adaptive Persistent Kernels","created":"2026-07-20","cited_by_count":1,"area":"moe"},{"id":"2410.07531","title":"Reducing the Cost of Dropout in Flash-Attention by Hiding RNG with GEMM","created":"2024-10-10","cited_by_count":0,"area":null},{"id":"2412.13614","title":"Reverse Region-to-Entity Annotation for Pixel-Level Visual Entity Linking","created":"2024-12-18","cited_by_count":0,"area":null},{"id":"2509.16702","title":"Animalbooth: multimodal feature enhancement for animal subject personalization","created":"2025-09-20","cited_by_count":0,"area":null},{"id":"2510.20111","title":"AsyncHZP: Hierarchical ZeRO Parallelism with Asynchronous Scheduling for Scalable LLM Training","created":"2025-10-23","cited_by_count":0,"area":"moe"},{"id":"2511.12500","title":"Iris: First-Class Multi-GPU Programming Experience in Triton","created":"2025-11-16","cited_by_count":0,"area":null},{"id":"2604.24013","title":"CommFuse: Hiding Tail Latency via Communication Decomposition and Fusion for Distributed LLM Training","created":"2026-04-27","cited_by_count":0,"area":null},{"id":"2605.21603","title":"DynaFlow: Transparent and Flexible Intra-Device Parallelism via Programmable Operator Scheduling","created":"2026-05-20","cited_by_count":0,"area":null},{"id":"2605.23764","title":"HyperParallel-MoE: Multi-Core Interleaved Scheduling for Fast MoE Training on Ascend NPUs","created":"2026-05-22","cited_by_count":0,"area":"moe"},{"id":"2606.09200","title":"Resource-aware Computation-Communication Overlap for multi-GPU ML Workloads","created":"2026-06-08","cited_by_count":0,"area":null},{"id":"2606.11169","title":"Piper: A Programmable Distributed Training System","created":"2026-06-09","cited_by_count":0,"area":null},{"id":"2607.07862","title":"CTA-Pipelining: A Latency-Oriented Spatial Scaling Method for Multi-GPU Systems","created":"2026-07-08","cited_by_count":0,"area":null},{"id":"2608.09160","title":"SwiftQK: Fast and Communication-Efficient Tensor Parallelism for Query-Key Normalization","created":"2026-08-10","cited_by_count":0,"area":null},{"id":"2609.07536","title":"Analytical Resource Management for Fine-grained MoE Computation-Communication Overlap","created":"2026-09-07","cited_by_count":0,"area":"moe"},{"id":"2609.13585","title":"mKernel: Fast Multi-GPU, Multi-Node Fused Kernels","created":"2026-09-11","cited_by_count":0,"area":null},{"id":"2609.13846","title":"Affinity-Aware Sharding for Delayed Tensor Parallelism","created":"2026-09-12","cited_by_count":0,"area":null},{"id":"2609.36070","title":"Mixture-of-Kittens: MoE Megakernel for NVL72s","created":"2026-09-28","cited_by_count":0,"area":"moe"},{"id":"2609.40093","title":"Efficient Expert-Parallel Communication on PCIe-Connected Consumer GPUs","created":"2026-09-30","cited_by_count":0,"area":"moe"},{"id":"2610.14277","title":"Echo: Error-Compensated 4-Bit Tensor-Parallel Communication for LLM Training and Inference","created":"2026-10-09","cited_by_count":0,"area":null},{"id":"2610.14935","title":"BSIPipe: Bidirectional Sequence-Level Interleaved Pipeline Parallelism for Long-Sequence Large Language Model Training","created":"2026-10-10","cited_by_count":0,"area":null},{"id":"2610.17901","title":"FlashMask-3: Efficient and Expressive Mask-Aware Distributed Attention","created":"2026-10-13","cited_by_count":0,"area":null}],"next_cursor":null}