mlsanityData to 14 Oct 2026

ORXE: Orchestrating Experts for Dynamically Configurable Efficiency

arXiv 2505.04850 ↗First posted 7 May 2025cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

This paper presents ORXE, a modular and adaptable framework for achieving real-time configurable efficiency in AI models. By leveraging a collection of pre-trained experts with diverse computational costs and performance levels, ORXE dynamically adjusts inference pathways based on the complexity of input samples. Unlike conventional approaches that require complex metamodel training, ORXE achieves high efficiency and flexibility without complicating the development process. The proposed system utilizes a confidence-based gating mechanism to allocate appropriate computational resources for each input. ORXE also supports adjustments to the preference between inference cost and prediction performance across a wide range during runtime. We implemented a training-free ORXE system for image classification tasks, evaluating its efficiency and accuracy across various devices. The results demonstrate that ORXE achieves superior performance compared to individual experts and other dynamic models in most cases. This approach can be extended to other applications, providing a scalable solution for diverse real-world deployment scenarios.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 55 of its 69 references that are in the map. First posted on arXiv 7 May 2025.

Builds on

55 of 69 references.

References · 55

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  4. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  5. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  6. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  7. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  8. 2020Scaling Laws for Neural Language Modelscited by 6,586
  9. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,442
  10. 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,509
  11. 2021Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsitycited by 3,073
  12. 2019Searching for MobileNetV3cited by 1,994
38 more
  1. 2019Do ImageNet Classifiers Generalize to ImageNet?cited by 1,515
  2. 2016Pruning Filters for Efficient ConvNetscited by 1,143
  3. 2019Learning Robust Global Representations by Penalizing Local Predictive Powercited by 1,013
  4. 2019Once-for-All: Train One Network and Specialize it for Efficient Deploymentcited by 787
  5. 2021Scaling Vision with Sparse Mixture of Expertscited by 747
  6. 2023ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoderscited by 712
  7. 2017BranchyNet: Fast Inference via Early Exiting from Deep Neural Networkscited by 596
  8. 2020What is the State of Neural Network Pruning?cited by 583
  9. 2021CoAtNet: Marrying Convolution and Attention for All Data Sizescited by 569
  10. 2022MaxViT: Multi-Axis Vision Transformercited by 476
  11. 2017SkipNet: Learning Dynamic Routing in Convolutional Networkscited by 332
  12. 2018Slimmable Neural Networkscited by 305
  13. 2022BEiT v2: Masked Image Modeling with Vector-Quantized Visual Tokenizerscited by 303
  14. 2023Hiera: A Hierarchical Vision Transformer without the Bells-and-Whistlescited by 290
  15. 2024Mixture-of-Depths: Dynamically allocating compute in transformer-based language modelscited by 258
  16. 2022Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpointscited by 209
  17. 2022Multimodal Contrastive Learning with LIMoE: the Language-Image Mixture of Expertscited by 203
  18. 2017BlockDrop: Dynamic Inference Paths in Residual Networkscited by 189
  19. 2022DaViT: Dual Attention Vision Transformerscited by 175
  20. 2024Scaling Laws for Fine-Grained Mixture of Expertscited by 173
  21. 2023EfficientViT: Memory Efficient Vision Transformer with Cascaded Group Attentioncited by 172
  22. 2017Convolutional Networks with Adaptive Inference Graphscited by 148
  23. 2020Resolution Adaptive Networks for Efficient Inferencecited by 146
  24. 2022Revisiting Neural Scaling Laws in Language and Visioncited by 132
  25. 2022MetaFormer Baselines for Visioncited by 132
  26. 2018Dynamic Channel Pruning: Feature Boosting and Suppressioncited by 129
  27. 2022Rethinking Vision Transformers for MobileNet Size and Speedcited by 121
  28. 2020Glance and Focus: a Dynamic Approach to Reducing Spatial Redundancy in Image Classificationcited by 100
  29. 2023MatFormer: Nested Transformer for Elastic Inferencecited by 85
  30. 2015Conditional Deep Learning for Energy-Efficient and Enhanced Pattern Recognitioncited by 67
  31. 2022A Comprehensive Survey on Model Quantization for Deep Neural Networks in Image Classificationcited by 65
  32. 2022Next-ViT: Next Generation Vision Transformer for Efficient Deployment in Realistic Industrial Scenarioscited by 65
  33. 2022CF-ViT: A General Coarse-to-Fine Method for Vision Transformercited by 63
  34. 2020A Panda? No, It's a Sloth: Slowdown Attacks on Adaptive Multi-Exit Neural Network Inferencecited by 61
  35. 2023Scaling Expert Language Models with Unsupervised Domain Discoverycited by 60
  36. 2022Learning to Weight Samples for Dynamic Early-exiting Networkscited by 44
  37. 2022EfficientViT: Multi-Scale Linear Attention for High-Resolution Dense Predictioncited by 42
  38. 2020The Cascade Transformer: an Application for Efficient Answer Sentence Selectioncited by 20

50 most cited of 55 All 55 (JSON)

Built on by

Not cited yet.