# ForkMerge: Mitigating Negative Transfer in Auxiliary-Task Learning

arXiv: https://arxiv.org/abs/2301.12618 · HTML: https://mlsanity.com/paper/2301.12618/

## Key facts

- Cited by 28 papers in the map (data through 14 Oct 2026).
- Builds on 60 of its 87 references that are in the map.
- First posted on arXiv 30 Jan 2023.

## Builds on (most cited first, 30 of 60)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks](https://mlsanity.com/paper/1703.03400/) (2017)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1802.02611/) (2018)
- [Wide Residual Networks](https://mlsanity.com/paper/1605.07146/) (2016)
- [FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence](https://mlsanity.com/paper/2001.07685/) (2020)
- [Unsupervised Representation Learning by Predicting Image Rotations](https://mlsanity.com/paper/1803.07728/) (2018)
- [Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour](https://mlsanity.com/paper/1706.02677/) (2017)
- [Big Self-Supervised Models are Strong Semi-Supervised Learners](https://mlsanity.com/paper/2006.10029/) (2020)
- [Progressive Neural Networks](https://mlsanity.com/paper/1606.04671/) (2016)
- [Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics](https://mlsanity.com/paper/1705.07115/) (2017)
- [Moment Matching for Multi-Source Domain Adaptation](https://mlsanity.com/paper/1812.01754/) (2018)
- [Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift](https://mlsanity.com/paper/1906.02530/) (2019)
- [Gradient Surgery for Multi-Task Learning](https://mlsanity.com/paper/2001.06782/) (2020)
- [Visualizing the Loss Landscape of Neural Nets](https://mlsanity.com/paper/1712.09913/) (2017)
- [Temporal Ensembling for Semi-Supervised Learning](https://mlsanity.com/paper/1610.02242/) (2016)
- [Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning](https://mlsanity.com/paper/1704.03976/) (2017)
- [GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks](https://mlsanity.com/paper/1711.02257/) (2017)
- [Multi-Task Learning as Multi-Objective Optimization](https://mlsanity.com/paper/1810.04650/) (2018)
- [Taskonomy: Disentangling Task Transfer Learning](https://mlsanity.com/paper/1804.08328/) (2018)
- [End-to-End Multi-Task Learning with Attention](https://mlsanity.com/paper/1803.10704/) (2018)
- [Cross-stitch Networks for Multi-task Learning](https://mlsanity.com/paper/1604.03539/) (2016)
- [Conflict-Averse Gradient Descent for Multi-task Learning](https://mlsanity.com/paper/2110.14048/) (2021)
- [Revisiting the Calibration of Modern Neural Networks](https://mlsanity.com/paper/2106.07998/) (2021)
- [Which Tasks Should Be Learned Together in Multi-task Learning?](https://mlsanity.com/paper/1905.07553/) (2019)

## Built on by (most cited first, 28 of 28)

- [Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities](https://mlsanity.com/paper/2408.07666/) (2024)
- [AdaMerging: Adaptive Model Merging for Multi-Task Learning](https://mlsanity.com/paper/2310.02575/) (2023)
- [Twin-Merging: Dynamic Integration of Modular Expertise in Model Merging](https://mlsanity.com/paper/2406.15479/) (2024)
- [Deep Model Fusion: A Survey](https://mlsanity.com/paper/2309.15698/) (2023)
- [Modeling Multi-Task Model Merging as Adaptive Projective Gradient Descent](https://mlsanity.com/paper/2501.01230/) (2025)
- [Selective Task Group Updates for Multi-Task Optimization](https://mlsanity.com/paper/2502.11986/) (2025)
- [StainDiffuser: MultiTask Dual Diffusion Model for Virtual Staining](https://mlsanity.com/paper/2403.11340/) (2024)
- [HENet++: Hybrid Encoding and Multi-task Learning for 3D Perception and End-to-end Autonomous Driving](https://mlsanity.com/paper/2511.07106/) (2025)
- [FW-Merging: Scaling Model Merging with Frank-Wolfe Optimization](https://mlsanity.com/paper/2503.12649/) (2025)
- [ADSNet: Cross-Domain LTV Prediction with an Adaptive Siamese Network in Advertising](https://mlsanity.com/paper/2406.10517/) (2024)
- [Large Language Model as Universal Retriever in Industrial-Scale Recommender System](https://mlsanity.com/paper/2502.03041/) (2025)
- [Resolving Token-Space Gradient Conflicts: Token Space Manipulation for Transformer-Based Multi-Task Learning](https://mlsanity.com/paper/2507.07485/) (2025)
- [Mitigate Negative Transfer with Similarity Heuristic Lifelong Prompt Tuning](https://mlsanity.com/paper/2406.12251/) (2024)
- [Pareto Continual Learning: Preference-Conditioned Learning and Adaption for Dynamic Stability-Plasticity Trade-off](https://mlsanity.com/paper/2503.23390/) (2025)
- [Exploring the Effectiveness and Consistency of Task Selection in Intermediate-Task Transfer Learning](https://mlsanity.com/paper/2407.16245/) (2024)
- [Can Optimization Trajectories Explain Multi-Task Transfer?](https://mlsanity.com/paper/2408.14677/) (2024)
- [TTVD: Towards a Geometric Framework for Test-Time Adaptation Based on Voronoi Diagram](https://mlsanity.com/paper/2412.07980/) (2024)
- [From Basic to Extra Features: Hypergraph Transformer Pretrain-then-Finetuning for Balanced Clinical Predictions on EHR](https://mlsanity.com/paper/2406.05682/) (2024)
- [Variational Model Merging for Pareto Front Estimation in Multitask Finetuning](https://mlsanity.com/paper/2412.08147/) (2024)
- [Graceful Forgetting in Generative Language Models](https://mlsanity.com/paper/2505.19715/) (2025)
- [Smooth-Distill: A Self-distillation Framework for Multitask Learning with Wearable Sensor Data](https://mlsanity.com/paper/2507.00061/) (2025)
- [Mitigating Negative Transfer via Reducing Environmental Disagreement](https://mlsanity.com/paper/2510.24044/) (2025)
- [Auxiliary Gene Learning: Spatial Gene Expression Estimation by Auxiliary Gene Selection](https://mlsanity.com/paper/2511.18336/) (2025)
- [Merging without Forgetting: Continual Fusion of Task-Specific Models via Optimal Transport](https://mlsanity.com/paper/2511.19561/) (2025)
- [A Graph Neural Network with Auxiliary Task Learning for Missing PMU Data Reconstruction](https://mlsanity.com/paper/2512.24542/) (2025)
- [Semi-overlapping Multi-bandit Best Arm Identification for Sequential Support Network Learning](https://mlsanity.com/paper/2512.24959/) (2025)
- [MoAKE: Toward Unified All-in-One Action Quality Assessment via Mixture of Action Knowledge Experts](https://mlsanity.com/paper/2607.19826/) (2026)
- [Generalist-Specialist Mixture-of-Experts for Rare Pathology Detection in Multimodal Imaging](https://mlsanity.com/paper/2609.18688/) (2026)

## Abstract (arXiv)

Auxiliary-Task Learning (ATL) aims to improve the performance of the target task by leveraging the knowledge obtained from related tasks. Occasionally, learning multiple tasks simultaneously results in lower accuracy than learning only the target task, which is known as negative transfer. This problem is often attributed to the gradient conflicts among tasks, and is frequently tackled by coordinating the task gradients in previous works. However, these optimization-based methods largely overlook the auxiliary-target generalization capability. To better understand the root cause of negative transfer, we experimentally investigate it from both optimization and generalization perspectives. Based on our findings, we introduce ForkMerge, a novel approach that periodically forks the model into multiple branches, automatically searches the varying task weights by minimizing target validation errors, and dynamically merges all branches to filter out detrimental task-parameter updates. On a series of auxiliary-task learning benchmarks, ForkMerge outperforms existing methods and effectively mitigates negative transfer.