# Janus: Collaborative Vision Transformer Under Dynamic Network Environment

arXiv: https://arxiv.org/abs/2502.10047 · HTML: https://mlsanity.com/paper/2502.10047/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 47 references that are in the map.
- First posted on arXiv 14 Feb 2025.

## Builds on (most cited first, 21 of 21)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [The Kinetics Human Action Video Dataset](https://mlsanity.com/paper/1705.06950/) (2017)
- [Token Merging: Your ViT But Faster](https://mlsanity.com/paper/2210.09461/) (2022)
- [Transformers in Vision: A Survey](https://mlsanity.com/paper/2101.01169/) (2021)
- [Exploring Plain Vision Transformer Backbones for Object Detection](https://mlsanity.com/paper/2203.16527/) (2022)
- [A Survey on Visual Transformer](https://mlsanity.com/paper/2012.12556/) (2020)
- [MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformer](https://mlsanity.com/paper/2110.02178/) (2021)
- [VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking](https://mlsanity.com/paper/2303.16727/) (2023)
- [Masked Autoencoders As Spatiotemporal Learners](https://mlsanity.com/paper/2205.09113/) (2022)
- [Adaptive Token Sampling For Efficient Vision Transformers](https://mlsanity.com/paper/2111.15667/) (2021)
- [PTQ4ViT: Post-training quantization for vision transformers with twin uniform quantization](https://mlsanity.com/paper/2111.12293/) (2021)
- [Evo-ViT: Slow-Fast Token Evolution for Dynamic Vision Transformer](https://mlsanity.com/paper/2108.01390/) (2021)
- [Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge Computing](https://mlsanity.com/paper/1910.05316/) (2019)
- [Patch Slimming for Efficient Vision Transformers](https://mlsanity.com/paper/2106.02852/) (2021)
- [IA-RED2: Interpretability-Aware Redundancy Reduction for Vision Transformers](https://mlsanity.com/paper/2106.12620/) (2021)
- [SPINN: Synergistic Progressive Inference of Neural Networks over Device and Cloud](https://mlsanity.com/paper/2008.06402/) (2020)
- [Semantic Segmentation using Vision Transformers: A survey](https://mlsanity.com/paper/2305.03273/) (2023)
- [Lite Vision Transformer with Enhanced Self-Attention](https://mlsanity.com/paper/2112.10809/) (2021)
- [Learning Efficient Vision Transformers via Fine-Grained Manifold Distillation](https://mlsanity.com/paper/2107.01378/) (2021)
- [PSAQ-ViT V2: Towards Accurate and General Data-Free Quantization for Vision Transformers](https://mlsanity.com/paper/2209.05687/) (2022)
- [ShiftAddNAS: Hardware-Inspired Search for More Accurate and Efficient Neural Networks](https://mlsanity.com/paper/2205.08119/) (2022)

## Built on by (most cited first, 4 of 4)

- [NuWa: Deriving Lightweight Class-Specific Vision Transformers for Edge Devices](https://mlsanity.com/paper/2504.03118/) (2025)
- [ViTMAlis: Towards Latency-Critical Mobile Video Analytics with Vision Transformers](https://mlsanity.com/paper/2601.21362/) (2026)
- [CST: Collaborative Selective Transmission for Communication-Efficient Multimodal Edge Inference](https://mlsanity.com/paper/2608.22115/) (2026)
- [Iapetus: Content-Aware Hierarchical Scheduling for Collaborative ViT Inference in LEO Satellite Networks](https://mlsanity.com/paper/2609.03318/) (2026)

## Abstract (arXiv)

Vision Transformers (ViTs) have outperformed traditional Convolutional Neural Network architectures and achieved state-of-the-art results in various computer vision tasks. Since ViTs are computationally expensive, the models either have to be pruned to run on resource-limited edge devices only or have to be executed on remote cloud servers after receiving the raw data transmitted over fluctuating networks. The resulting degraded performance or high latency all hinder their widespread applications. In this paper, we present Janus, the first framework for low-latency cloud-device collaborative Vision Transformer inference over dynamic networks. Janus overcomes the intrinsic model limitations of ViTs and realizes collaboratively executing ViT models on both cloud and edge devices, achieving low latency, high accuracy, and low communication overhead. Specifically, Janus judiciously combines token pruning techniques with a carefully designed fine-to-coarse model splitting policy and non-static mixed pruning policy. It attains a balance between accuracy and latency by dynamically selecting the optimal pruning level and split point. Experimental results across various tasks demonstrate that Janus enhances throughput by up to 5.15 times and reduces latency violation ratios by up to 98.7% when compared with baseline approaches under various network environments.