Distributed Training papers, 2023–2026

Key findings

• Distributed Training went from 95 papers in 2023 to 98 in 2026, roughly 1.0x in three years.
• NeurIPS publishes the most Distributed Training work: 208 papers from 2023 to 2026, 53% of the four-venue total.
• The fastest-growing overlap is with Multimodal LLMs: papers tagged with both rose from 1 in 2023 to 10 in 2026.
• Among the 5 topics under Efficiency and Systems, it ranks #4 by 2026 paper count; the fastest grower in the group, Hardware-Aware ML, is up 4.4x versus 1.0x here.

Representative papers

Spread across years and venues so every edition shows up

• NeurIPS 2026AgentWeave: Efficient Distributed Agent Serving via Flow Decomposition
• ICML 2026Breaking the Capacity Bottleneck in Model-Heterogeneous Federated Learning via Gradual Model Restoration
• ICLR 2026Cascadia: An Efficient Cascade Serving System for Large Language Models
• CVPR 2026Finding Distributed Object-Centric Properties in Self-Supervised Transformers
• NeurIPS 2025A Fair Federated Learning Method for Handling Client Participation Probability Inconsistencies in Heterogeneous Environments
• ICML 2025BSemiFL: Semi-supervised Federated Learning via a Bayesian Approach
• ICLR 2025Addressing Label Shift in Distributed Learning via Entropy Regularization
• CVPR 2025Real-time High-fidelity Gaussian Human Avatars with Position-based Interpolation of Spatially Distributed MLPs
• NeurIPS 2024Communication Efficient Distributed Training with Distributed Lion
• ICML 2024Embarrassingly Parallel GFlowNets
• ICLR 2024Benchmarking Algorithms for Federated Domain Generalization
• CVPR 2024Deciphering 'What' and 'Where' Visual Pathways from Spectral Clustering of Layer-Distributed Neural Representations
• NeurIPS 20232Direction: Theoretically Faster Distributed Training with Bidirectional Communication Compression
• ICML 2023Block Subsampled Randomized Hadamard Transform for Nyström Approximation on Distributed Architectures
• ICLR 2023Distributed Extra-gradient with Optimal Complexity and Communication Guarantees
• CVPR 2023DisCo-CLIP: A Distributed Contrastive Loss for Memory Efficient CLIP Training
• NeurIPS 2026Do We Need Asynchronous SGD? On the Near-Optimality of Synchronous Methods
• ICML 2026Clipping Makes Distributed and Federated Asynchronous SGD Robust to Stragglers
• ICLR 2026DualMap: Enabling Both Cache Affinity and Load Balancing for Distributed LLM Serving
• CVPR 2026Parallel Rigidity Matters for Bundle Adjustment

All Distributed Training papers

396 papers, page 2 of 8

51. ICML 2026Matrix-Free GPU Semidefinite Programming for Quantum Ordered Search at the k=6 Frontier
52. ICML 2026Mitigating Staleness in Asynchronous Pipeline Parallelism via Basis Rotation
53. ICML 2026MuLoCo: Muon is a Practical Inner Optimizer for DiLoCo
54. ICML 2026On the Role of Batch Size in Stochastic Conditional Gradient Methods
55. ICML 2026One-Step Gradient Delay is Not a Barrier for Large-Scale Asynchronous Pipeline Parallel LLM Pretraining
56. ICML 2026OServe: Accelerating LLM Serving via Spatial-Temporal Workload Orchestration
57. ICML 2026Random Scaling of Emergent Capabilities
58. ICML 2026Robust Federated Learning Against Adaptive Compression
59. ICML 2026Row-Stochastic Matrices Can Provably Outperform Doubly Stochastic Matrices in Decentralized Learning
60. ICML 2026scDataset: Scalable Data Loading for Deep Learning on Large-Scale Single-Cell Omics
61. ICML 2026SFedPO: Streaming Federated Learning with a Prediction Oracle under Temporal Shifts
62. ICML 2026Tight Stability Bounds for Robust Distributed Learning: Byzantine Failures Hurt Generalization More than Data Poisoning
63. NeurIPS 2026Adversary-Robust Learning from Fully Asynchronous Directional Derivative Estimates
64. NeurIPS 2026AgentWeave: Efficient Distributed Agent Serving via Flow Decomposition
65. NeurIPS 2026BReD: Block Replay Dithering for Stable Low-Bit EMA Optimizer States
66. NeurIPS 2026Can Model Merging Improve Aggregation in DiLoCo?
67. NeurIPS 2026Clapping: Removing Per-sample Storage for Pipeline Parallel Learning with Communication Compression
68. NeurIPS 2026CRAFT: Conflict-Resolved Aggregation for Federated Training
69. NeurIPS 2026DDBench: A Benchmark for Agentic Debugging on Distributed Systems
70. NeurIPS 2026Delightful Distributed Policy Gradient
71. NeurIPS 2026Dependency-Guided Parallel Decoding in Discrete Diffusion Language Models
72. NeurIPS 2026DISTMOE: Rehearsal-free Routing in Mixture-of-Experts for Distributed Instruction Tuning
73. NeurIPS 2026Do We Need Asynchronous SGD? On the Near-Optimality of Synchronous Methods
74. NeurIPS 2026ECHO-2: A Large-Scale Distributed Rollout Framework for Cost-Efficient Reinforcement Learning
75. NeurIPS 2026Efficient Algorithms for Distributed Saddle Problems
76. NeurIPS 2026Federated Concept-Based Models: Interpretable models with distributed supervision
77. NeurIPS 2026FedPeel: Peeling Stabilized Layers for Robust Heterogeneous Federated Learning
78. NeurIPS 2026FlashMask-3: Efficient and Expressive Mask-Aware Distributed Attention
79. NeurIPS 2026Focus on Where You Aggregate: Restricted SAM for Non-IID Federated Learning
80. NeurIPS 2026FRESCO: A Novel Consistency Control for Asynchronous Pipeline Parallel Training
81. NeurIPS 2026Fully Distributed Tâtonnement for Chores Markets
82. NeurIPS 2026GPU-Accelerated Synthesis of Mixed-Boolean Arithmetic: Beyond Caching
83. NeurIPS 2026HetCCL: Efficient LLM Training on Heterogeneous Vendor GPUs
84. NeurIPS 2026In STeP: Speculative Tensor Parallelism for Concurrent Heterogeneous Inference of LLMs
85. NeurIPS 2026LOSCAR-SGD: Local SGD with Communication-Computation Overlap and Delay-Corrected Sparse Model Averaging
86. NeurIPS 2026Measuring Layer-wise Intrinsic Dimensionality of FFNs in LLMs via PCA
87. NeurIPS 2026Mitigating Data Heterogeneity Effect in Client-Reshuffling-Based Federated Learning
88. NeurIPS 2026nnTrace: Detecting and Localizing Silent Bugs in Distributed Training
89. NeurIPS 2026On the Parallel Optimality of Exponentiated Gradient Descent
90. NeurIPS 2026P-EAGLE: Parallel-Drafting EAGLE with Scalable Training
91. NeurIPS 2026Principled Federated Random Forests for Heterogeneous Data
92. NeurIPS 2026Privacy Meets Hierarchy: Differentially Private Distributed Trilevel Learning
93. NeurIPS 2026ReCoVer: Resilient LLM Pre-Training System via Fault-Tolerant Collective and Versatile Workload
94. NeurIPS 2026Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneity
95. NeurIPS 2026Scalable Distributed Stochastic Optimization via Bidirectional Compression: Beyond Pessimistic Limits
96. NeurIPS 2026SMMBench: A Benchmark for Source-Distributed Multimodal Agent Memory
97. NeurIPS 2026vExpert: Virtualizing Expert Storage for Adaptive Load Balancing in Distributed MoE Inference
98. NeurIPS 2026When Must AI Training Stage Checkpoints? A Distributional Model of Durability Boundaries at Scale
99. CVPR 2025Easy-editable Image Vectorization with Multi-layer Multi-scale Distributed Visual Feature Embedding
100. CVPR 2025Real-time High-fidelity Gaussian Human Avatars with Position-based Interpolation of Spatially Distributed MLPs

FAQ

How many Distributed Training papers were published at NeurIPS, ICLR, ICML and CVPR from 2023 to 2026?

396 in total: 95 in 2023, 74 in 2024, 129 in 2025, 98 in 2026.

How many Distributed Training papers are at NeurIPS 2026?

NeurIPS 2026 has 36 Distributed Training papers, versus 69 the year before.

How many Distributed Training papers are at ICLR 2026?

ICLR 2026 has 18 Distributed Training papers, versus 18 the year before.

How many Distributed Training papers are at ICML 2026?

ICML 2026 has 42 Distributed Training papers, versus 40 the year before.

How many Distributed Training papers are at CVPR 2026?

CVPR 2026 has 2 Distributed Training papers, versus 2 the year before.

Which conference publishes the most Distributed Training papers?

NeurIPS, with 208 papers from 2023 to 2026.