mlsanityData to 7 Oct 2026

FAAR: Efficient Frequency-Aware Multi-Task Fine-Tuning via Automatic Rank Selection

arXiv 2603.20403 ↗First posted 20 Mar 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Adapting models pre-trained on large-scale datasets is a proven way to reach strong performance quickly for down-stream tasks. However, the growth of state-of-the-art mod-els makes traditional full fine-tuning unsuitable and difficult, especially for multi-task learning (MTL) where cost scales with the number of tasks. As a result, recent studies investigate parameter-efficient fine-tuning (PEFT) using low-rank adaptation to significantly reduce the number of trainable parameters. However, these existing methods use a single, fixed rank, which may not be optimal for differ-ent tasks or positions in the MTL architecture. Moreover, these methods fail to learn spatial information that cap-tures inter-task relationships and helps to improve diverse task predictions. This paper introduces Frequency-Aware and Automatic Rank (FAAR) for efficient MTL fine-tuning. Our method introduces Performance-Driven Rank Shrink-ing (PDRS) to allocate the optimal rank per adapter location and per task. Moreover, by analyzing the image frequency spectrum, FAAR proposes a Task-Spectral Pyramidal Decoder (TS-PD) that injects input-specific context into spatial bias learning to better reflect cross-task relationships. Experiments performed on dense visual task benchmarks show the superiority of our method in terms of both accuracy and efficiency compared to other PEFT methods in MTL. FAAR reduces the number of parameters by up to 9 times compared to traditional MTL fine-tuning whilst improving overall performance. Our code is available.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 31 of its 53 references that are in the map. First posted on arXiv 20 Mar 2026.

Builds on

31 of 53 references.

References · 31

  1. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,769
  2. 2019Deep High-Resolution Representation Learning for Human Pose Estimationcited by 1,339
  3. 2021BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-modelscited by 1,234
  4. 2021Towards a Unified View of Parameter-Efficient Transfer Learningcited by 863
  5. 2024DoRA: Weight-Decomposed Low-Rank Adaptationcited by 740
  6. 2021Compacter: Efficient Low-Rank Hypercomplex Adapter Layerscited by 458
  7. 2019Which Tasks Should Be Learned Together in Multi-task Learning?cited by 367
  8. 2023AdaLoRA: Adaptive Budget Allocation for Parameter-Efficient Fine-Tuningcited by 337
  9. 2021VL-Adapter: Parameter-Efficient Transfer Learning for Vision-and-Language Taskscited by 319
  10. 2021Parameter-efficient Multi-task Fine-tuning for Transformers via Shared Hypernetworkscited by 278
  11. 2022DyLoRA: Parameter Efficient Tuning of Pre-trained Models using Dynamic Search-Free Low-Rank Adaptationcited by 229
  12. 2018PAD-Net: Multi-Tasks Guided Prediction-and-Distillation Network for Simultaneous Depth Estimation and Scene Parsingcited by 224
19 more
  1. 2024Rethinking Inductive Biases for Surface Normal Estimationcited by 122
  2. 2023VMT-Adapter: Parameter-Efficient Transfer Learning for Multi-Task Dense Scene Understandingcited by 62
  3. 2024Multi-Task Dense Prediction via Mixture of Low-Rank Expertscited by 58
  4. 2022Polyhistor: Parameter-Efficient Multi-Task Adaptation for Dense Vision Taskscited by 52
  5. 2023TaskExpert: Dynamically Assembling Multi-Task Representations with Memorial Mixture-of-Expertscited by 45
  6. 2024MTLoRA: A Low-Rank Adaptation Approach for Efficient Multi-Task Learningcited by 44
  7. 2021Single Image Depth Prediction with Wavelet Decompositioncited by 34
  8. 2024AutoLoRA: Automatically Tuning Matrix Ranks in Low-Rank Adaptation Based on Meta Learningcited by 30
  9. 2024Image Compression for Machine and Human Vision with Spatial-Frequency Adaptationcited by 26
  10. 2024Parameter-Efficient Fine-Tuning for Pre-Trained Vision Models: A Survey and Benchmarkcited by 21
  11. 2023Adapter is All You Need for Tuning Visual Taskscited by 19
  12. 2025Frequency Dynamic Convolution for Dense Image Predictioncited by 19
  13. 2024On the Surprising Effectiveness of Attention Transfer for Vision Transformerscited by 16
  14. 2025A Survey on Parameter-Efficient Fine-Tuning for Foundation Models in Federated Learningcited by 15
  15. 2023Efficient Computation Sharing for Multi-Task Visual Scene Understandingcited by 8
  16. 2026A Real-Time Error Prevention System for Gaze-Based Interaction in Virtual Reality Based on Anomaly Detectioncited by 8
  17. 2025TADFormer : Task-Adaptive Dynamic Transformer for Efficient Multi-Task Learningcited by 5
  18. 2024MacFormer: Semantic Segmentation with Fine Object Boundariescited by 3
  19. 2024Optimizing Dense Visual Predictions Through Multi-Task Coherence and Prioritizationcited by 1

Built on by

Not cited yet.