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

208 papers, page 1 of 5

1. NeurIPS 2026Adversary-Robust Learning from Fully Asynchronous Directional Derivative Estimates
2. NeurIPS 2026AgentWeave: Efficient Distributed Agent Serving via Flow Decomposition
3. NeurIPS 2026BReD: Block Replay Dithering for Stable Low-Bit EMA Optimizer States
4. NeurIPS 2026Can Model Merging Improve Aggregation in DiLoCo?
5. NeurIPS 2026Clapping: Removing Per-sample Storage for Pipeline Parallel Learning with Communication Compression
6. NeurIPS 2026CRAFT: Conflict-Resolved Aggregation for Federated Training
7. NeurIPS 2026DDBench: A Benchmark for Agentic Debugging on Distributed Systems
8. NeurIPS 2026Delightful Distributed Policy Gradient
9. NeurIPS 2026Dependency-Guided Parallel Decoding in Discrete Diffusion Language Models
10. NeurIPS 2026DISTMOE: Rehearsal-free Routing in Mixture-of-Experts for Distributed Instruction Tuning
11. NeurIPS 2026Do We Need Asynchronous SGD? On the Near-Optimality of Synchronous Methods
12. NeurIPS 2026ECHO-2: A Large-Scale Distributed Rollout Framework for Cost-Efficient Reinforcement Learning
13. NeurIPS 2026Efficient Algorithms for Distributed Saddle Problems
14. NeurIPS 2026Federated Concept-Based Models: Interpretable models with distributed supervision
15. NeurIPS 2026FedPeel: Peeling Stabilized Layers for Robust Heterogeneous Federated Learning
16. NeurIPS 2026FlashMask-3: Efficient and Expressive Mask-Aware Distributed Attention
17. NeurIPS 2026Focus on Where You Aggregate: Restricted SAM for Non-IID Federated Learning
18. NeurIPS 2026FRESCO: A Novel Consistency Control for Asynchronous Pipeline Parallel Training
19. NeurIPS 2026Fully Distributed Tâtonnement for Chores Markets
20. NeurIPS 2026GPU-Accelerated Synthesis of Mixed-Boolean Arithmetic: Beyond Caching
21. NeurIPS 2026HetCCL: Efficient LLM Training on Heterogeneous Vendor GPUs
22. NeurIPS 2026In STeP: Speculative Tensor Parallelism for Concurrent Heterogeneous Inference of LLMs
23. NeurIPS 2026LOSCAR-SGD: Local SGD with Communication-Computation Overlap and Delay-Corrected Sparse Model Averaging
24. NeurIPS 2026Measuring Layer-wise Intrinsic Dimensionality of FFNs in LLMs via PCA
25. NeurIPS 2026Mitigating Data Heterogeneity Effect in Client-Reshuffling-Based Federated Learning
26. NeurIPS 2026nnTrace: Detecting and Localizing Silent Bugs in Distributed Training
27. NeurIPS 2026On the Parallel Optimality of Exponentiated Gradient Descent
28. NeurIPS 2026P-EAGLE: Parallel-Drafting EAGLE with Scalable Training
29. NeurIPS 2026Principled Federated Random Forests for Heterogeneous Data
30. NeurIPS 2026Privacy Meets Hierarchy: Differentially Private Distributed Trilevel Learning
31. NeurIPS 2026ReCoVer: Resilient LLM Pre-Training System via Fault-Tolerant Collective and Versatile Workload
32. NeurIPS 2026Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneity
33. NeurIPS 2026Scalable Distributed Stochastic Optimization via Bidirectional Compression: Beyond Pessimistic Limits
34. NeurIPS 2026SMMBench: A Benchmark for Source-Distributed Multimodal Agent Memory
35. NeurIPS 2026vExpert: Virtualizing Expert Storage for Adaptive Load Balancing in Distributed MoE Inference
36. NeurIPS 2026When Must AI Training Stage Checkpoints? A Distributional Model of Durability Boundaries at Scale
37. NeurIPS 2025A Fair Federated Learning Method for Handling Client Participation Probability Inconsistencies in Heterogeneous Environments
38. NeurIPS 2025Accelerating Parallel Diffusion Model Serving with Residual Compression
39. NeurIPS 2025ACCO: Accumulate While You Communicate for Communication-Overlapped Sharded LLM Training
40. NeurIPS 2025Agentic Plan Caching: Test-Time Memory for Fast and Cost-Efficient LLM Agents
41. NeurIPS 2025Approximate Gradient Coding for Distributed Learning with Heterogeneous Stragglers
42. NeurIPS 2025AREAL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning
43. NeurIPS 2025ASDSV: Multimodal Generation Made Efficient with Approximate Speculative Diffusion and Speculative Verification
44. NeurIPS 2025Block-Diagonal LoRA for Eliminating Communication Overhead in Tensor Parallel LoRA Serving
45. NeurIPS 2025Channel Simulation and Distributed Compression with Ensemble Rejection Sampling
46. NeurIPS 2025ClusterFusion: Expanding Operator Fusion Scope for LLM Inference via Cluster-Level Collective Primitive
47. NeurIPS 2025Communication-Efficient Diffusion Denoising Parallelization via Reuse-then-Predict Mechanism
48. NeurIPS 2025Communication-Efficient Language Model Training Scales Reliably and Robustly: Scaling Laws for DiLoCo
49. NeurIPS 2025Contextual Tokenization for Graph Inverted Indices
50. NeurIPS 2025Cost-Efficient LLM Training with Lifetime-Aware Tensor Offloading via GPUDirect Storage

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.