mlsanityData to 7 Oct 2026

Cross-stitch Networks for Multi-task Learning

arXiv 1604.03539 ↗First posted 12 Apr 2016cs.CV cs.LGcited by 466

Not on the map. Why?

Abstract (arXiv)

Multi-task learning in Convolutional Networks has displayed remarkable success in the field of recognition. This success can be largely attributed to learning shared representations from multiple supervisory tasks. However, existing multi-task approaches rely on enumerating multiple network architectures specific to the tasks at hand, that do not generalize. In this paper, we propose a principled approach to learn shared representations in ConvNets using multi-task learning. Specifically, we propose a new sharing unit: "cross-stitch" unit. These units combine the activations from multiple networks and can be trained end-to-end. A network with cross-stitch units can learn an optimal combination of shared and task-specific representations. Our proposed method generalizes across multiple tasks and shows dramatically improved performance over baseline methods for categories with few training examples.

Key facts. Cited by 466 papers in the map (data through 7 Oct 2026). First posted on arXiv 12 Apr 2016.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2019: 1; 2020: 1; 2021: 92; 2022: 86; 2023: 107; 2024: 82; 2025: 62; 2026: 35.

466 papers.

Sort:

By area:

  1. 2021Perceiver IO: A General Architecture for Structured Inputs & Outputscited by 616
  2. 2021Conflict-Averse Gradient Descent for Multi-task Learningcited by 441
  3. 2021Omnidata: A Scalable Pipeline for Making Multi-Task Mid-Level Vision Datasets from 3D Scanscited by 336
  4. 2023Blind Image Quality Assessment via Vision-Language Correspondence: A Multitask Learning Perspectivecited by 276
  5. 2023AdaMerging: Adaptive Model Merging for Multi-Task Learningcited by 248
  6. 2021MT-Opt: Continuous Multi-Task Robotic Reinforcement Learning at Scalecited by 226
  7. 2021Efficiently Identifying Task Groupings for Multi-Task Learningcited by 218
  8. 2022Omnivore: A Single Model for Many Visual Modalitiescited by 214
  9. 2022Multi-Task Learning as a Bargaining Gamecited by 209
  10. 2022BEVerse: Unified Perception and Prediction in Birds-Eye-View for Vision-Centric Autonomous Drivingcited by 191
  11. 2023Multitask Prompt Tuning Enables Parameter-Efficient Transfer Learningcited by 115
  12. 2022Federated Learning with Partial Model Personalizationcited by 114
38 more
  1. 2023Task-Specific Skill Localization in Fine-tuned Language Modelscited by 104
  2. 2021Reasonable Effectiveness of Random Weighting: A Litmus Test for Multi-Task Learningcited by 100
  3. 2022M3ViT: Mixture-of-Experts Vision Transformer for Efficient Multi-task Learning with Model-Accelerator Co-designcited by 97
  4. 2024Representation Surgery for Multi-Task Model Mergingcited by 95
  5. 2022InvPT: Inverted Pyramid Multi-task Transformer for Dense Scene Understandingcited by 89
  6. 2021RotoGrad: Gradient Homogenization in Multitask Learningcited by 82
  7. 2023Modular Deep Learningcited by 82
  8. 2021One Model to Serve All: Star Topology Adaptive Recommender for Multi-Domain CTR Predictioncited by 78
  9. 2021On Inductive Biases for Heterogeneous Treatment Effect Estimationcited by 74
  10. 2023FAMO: Fast Adaptive Multitask Optimizationcited by 74
  11. 2022Auto-Lambda: Disentangling Dynamic Task Relationshipscited by 73
  12. 2021Exploring Relational Context for Multi-Task Dense Predictioncited by 71
  13. 2023Independent Component Alignment for Multi-Task Learningcited by 69
  14. 2023vMAP: Vectorised Object Mapping for Neural Field SLAMcited by 67
  15. 2022In Defense of the Unitary Scalarization for Deep Multi-Task Learningcited by 65
  16. 2023Advances and Challenges in Meta-Learning: A Technical Reviewcited by 65
  17. 2021OTCE: A Transferability Metric for Cross-Domain Cross-Task Representationscited by 62
  18. 2021Multi-Task Learning in Natural Language Processing: An Overviewcited by 61
  19. 2023No Representation Rules Them All in Category Discoverycited by 61
  20. 2021PolyViT: Co-training Vision Transformers on Images, Videos and Audiocited by 58
  21. 2024Multi-Task Dense Prediction via Mixture of Low-Rank Expertscited by 58
  22. 2022Disentangling Visual Embeddings for Attributes and Objectscited by 55
  23. 2021Model Zoo: A Growing "Brain" That Learns Continuallycited by 53
  24. 2024MetaGPT: Merging Large Language Models Using Model Exclusive Task Arithmeticcited by 53
  25. 20244M-21: An Any-to-Any Vision Model for Tens of Tasks and Modalitiescited by 52
  26. 2024A Survey on Model MoErging: Recycling and Routing Among Specialized Experts for Collaborative Learningcited by 52
  27. 2023Revisiting Scalarization in Multi-Task Learning: A Theoretical Perspectivecited by 51
  28. 2023Recon: Reducing Conflicting Gradients from the Root for Multi-Task Learningcited by 50
  29. 2024LoRA Soups: Merging LoRAs for Practical Skill Composition Taskscited by 49
  30. 2025TrackVLA: Embodied Visual Tracking in the Wildcited by 46
  31. 2023TaskExpert: Dynamically Assembling Multi-Task Representations with Memorial Mixture-of-Expertscited by 45
  32. 2024MTLoRA: A Low-Rank Adaptation Approach for Efficient Multi-Task Learningcited by 44
  33. 2024DINO-Foresight: Looking into the Future with DINOcited by 44
  34. 2021Learning Multiple Dense Prediction Tasks from Partially Annotated Datacited by 42
  35. 2022Task Adaptive Parameter Sharing for Multi-Task Learningcited by 42
  36. 2022KG-SP: Knowledge Guided Simple Primitives for Open World Compositional Zero-Shot Learningcited by 41
  37. 2022Hypernetworks in Meta-Reinforcement Learningcited by 39
  38. 2022One Model is All You Need: Multi-Task Learning Enables Simultaneous Histology Image Segmentation and Classificationcited by 38

50 most cited of 466 All 466 (JSON)