TSFLora: Token-Compressed Split Fine-Tuning for Wireless Edge Networks

• URL: http://arxiv.org/abs/2605.23988v1
• Date: Sun, 17 May 2026 08:50:01 GMT
• Title: TSFLora: Token-Compressed Split Fine-Tuning for Wireless Edge Networks
• Abstract summary: TSFLora is a token-compressed split fine-tuning framework for communication-efficient LAM adaptation at the edge.<n> TSFLora combines attention-guided token selection, token merging, low-bit activation quantization, and LoRA-based adaptation within a split federated training pipeline.
• Score: 41.95924024036396
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Adapting large AI models (LAMs) to personalized edge data is challenging because wireless devices have limited memory, computation, and uplink capacity. Federated fine-tuning preserves data privacy but still requires each device to host the full model, while split learning reduces device memory at the cost of heavy activation transmission. This paper proposes TSFLora, a token-compressed split fine-tuning framework for communication-efficient LAM adaptation at the edge. TSFLora combines attention-guided token selection, token merging, low-bit activation quantization, and LoRA-based adaptation within a split federated training pipeline. The key idea is to compress the intermediate token sequence before transmission so that the system reduces both uplink traffic and server-side processing without changing the frozen backbone. Experiments on ViT models over CIFAR-10, CIFAR-100, and TinyImageNet show that TSFLora achieves up to \textbf{6.8$\times$} communication reduction and \textbf{41\%} memory saving while maintaining competitive accuracy.

Related papers

• Profiling-Driven Adaptive Distributed Transformer Inference on Embedded Edge Deployment [60.442064966340524]
  We present a prototype study on NVIDIA Jetson Orin Nano devices connected over WiFi.<n>Key finding is that the dominant bottleneck is not just network bandwidth but also the CPU- GPU staging during communication.<n>Experiments show that this strategy reduces latency by 65%-77% and energy consumption by 34%-52% relative to full-tensor exchange.
  arXiv  Detail & Related papers  (2026-05-25T10:39:28Z)
• FED-FSTQ: Fisher-Guided Token Quantization for Communication-Efficient Federated Fine-Tuning of LLMs on Edge Devices [12.523170571488711]
  Federated fine-tuning provides a practical route to adapt large language models (LLMs) on edge devices without centralizing private data.<n>We propose Fed-FSTQ, a Fisher-guided token quantization system primitive for communication-efficient federated LLM fine-tuning.<n>We show that Fed-FSTQ reduces cumulative uplink traffic required to reach a fixed quality threshold by 46x relative to a standard LoRA baseline, and improves end-to-end wall-clock time-to-accuracy by 52%.
  arXiv  Detail & Related papers  (2026-04-28T09:29:41Z)
• Privacy-Preserving Federated Action Recognition via Differentially Private Selective Tuning and Efficient Communication [25.243649529257052]
  Federated video action recognition enables collaborative model training without sharing raw video data.<n>We propose textitFedDP-STECAR framework, which selectively fine-tunes and perturbs task-relevant layers under Differential Privacy (DP)<n>By transmitting only the tuned layers during aggregation, communication traffic is reduced by over 99% compared to full-model updates.
  arXiv  Detail & Related papers  (2026-03-22T16:05:14Z)
• Why Should the Server Do It All?: A Scalable, Versatile, and Model-Agnostic Framework for Server-Light DNN Inference over Massively Distributed Clients via Training-Free Intermediate Feature Compression [6.932768187544348]
  We introduce SLICER, a retraining-free, architecture-agnostic framework that compresses IFs to reduce both communication and server load in split computing.<n>Across standard vision and LLM workloads, SLICER reduces uplink volume by up to 10x and server GPU time by up to 4.4x.
  arXiv  Detail & Related papers  (2025-11-03T08:44:13Z)
• Federated Split Learning with Improved Communication and Storage Efficiency [9.277834710955766]
  Federated learning (FL) is one of the popular distributed machine learning (ML) solutions but incurs significant communication and computation costs at edge devices.<n>This paper proposes a novel communication and storage efficient federated split learning method, CSE-FSL, which utilizes an auxiliary network to locally update the clients while keeping a single model at the server.
  arXiv  Detail & Related papers  (2025-07-21T17:21:16Z)
• Model Parallelism With Subnetwork Data Parallelism [21.914077370806016]
  Pre-training large neural networks at scale imposes heavy memory demands on accelerators and often requires costly communication.<n>We introduce Subnetwork Data Parallelism (SDP), a distributed training framework that partitions a model into structuredworks trained across workers without exchanging activations.<n>We study two complementary masking regimes: backward masking, which applies sparsity only in the backward step to retain unbiased gradients, and forward masking, which also removes parameters in the forward pass to deliver stronger efficiency gains.
  arXiv  Detail & Related papers  (2025-07-11T21:25:11Z)
• Sparse-Tuning: Adapting Vision Transformers with Efficient Fine-tuning and Inference [14.030836300221756]
  textbfSparse-Tuning is a novel PEFT method that accounts for the information redundancy in images and videos. Sparse-Tuning minimizes the quantity of tokens processed at each layer, leading to a quadratic reduction in computational and memory overhead. Our results show that our Sparse-Tuning reduces GFLOPs to textbf62%-70% of the original ViT-B while achieving state-of-the-art performance.
  arXiv  Detail & Related papers  (2024-05-23T15:34:53Z)
• Efficient Asynchronous Federated Learning with Sparsification and Quantization [55.6801207905772]
  Federated Learning (FL) is attracting more and more attention to collaboratively train a machine learning model without transferring raw data. FL generally exploits a parameter server and a large number of edge devices during the whole process of the model training. We propose TEASQ-Fed to exploit edge devices to asynchronously participate in the training process by actively applying for tasks.
  arXiv  Detail & Related papers  (2023-12-23T07:47:07Z)
• Adaptive Federated Pruning in Hierarchical Wireless Networks [69.6417645730093]
  Federated Learning (FL) is a privacy-preserving distributed learning framework where a server aggregates models updated by multiple devices without accessing their private datasets. In this paper, we introduce model pruning for HFL in wireless networks to reduce the neural network scale. We show that our proposed HFL with model pruning achieves similar learning accuracy compared with the HFL without model pruning and reduces about 50 percent communication cost.
  arXiv  Detail & Related papers  (2023-05-15T22:04:49Z)
• An Adaptive Device-Edge Co-Inference Framework Based on Soft Actor-Critic [72.35307086274912]
  High-dimension parameter model and large-scale mathematical calculation restrict execution efficiency, especially for Internet of Things (IoT) devices. We propose a new Deep Reinforcement Learning (DRL)-Soft Actor Critic for discrete (SAC-d), which generates the emphexit point, emphexit point, and emphcompressing bits by soft policy iterations. Based on the latency and accuracy aware reward design, such an computation can well adapt to the complex environment like dynamic wireless channel and arbitrary processing, and is capable of supporting the 5G URL
  arXiv  Detail & Related papers  (2022-01-09T09:31:50Z)
• A Joint Energy and Latency Framework for Transfer Learning over 5G Industrial Edge Networks [53.26338041079138]
  We propose a transfer learning-enabled edge-CNN framework for 5G industrial edge networks. In particular, the edge server can use the existing image dataset to train the CNN in advance. With the aid of TL, the devices that are not participating in the training only need to fine-tune the trained edge-CNN model without training from scratch.
  arXiv  Detail & Related papers  (2021-04-19T15:13:16Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.