English
Related papers

Related papers: Subspace-Boosted Model Merging

200 papers

Multi-task learning improves generalization, but when does it reduce the model capacity required to learn? We provide a systematic study of how joint training affects the learning transition, the minimum model size at which a task can be…

Machine Learning · Computer Science 2026-02-03 Csaba Both , Benjamin Hoover , Hendrik Strobelt , Dmitry Krotov , Daniel Karl I. Weidele , Mauro Martino , Nima Dehmamy

Ensembles of machine learning models have been well established as a powerful method of improving performance over a single model. Traditionally, ensembling algorithms train their base learners independently or sequentially with the goal of…

Machine Learning · Computer Science 2023-11-01 Alan Jeffares , Tennison Liu , Jonathan Crabbé , Mihaela van der Schaar

Mixture of Experts (MoE) models have become central to scaling large language models, yet their mechanistic differences from dense networks remain poorly understood. Previous work has explored how dense models use \textit{superposition} to…

Machine Learning · Computer Science 2025-12-29 Marmik Chaudhari , Jeremi Nuer , Rome Thorstenson

Although multi-task learning is widely applied in intelligent services, traditional multi-task modeling methods often require customized designs based on specific task combinations, resulting in a cumbersome modeling process. Inspired by…

Machine Learning · Computer Science 2025-04-15 Jingxuan Zhou , Weidong Bao , Ji Wang , Zhengyi Zhong , Dayu Zhang

Mixture-of-Experts models enable large language models to scale efficiently, as they only activate a subset of experts for each input. Their core mechanisms, Top-k routing and auxiliary load balancing, remain heuristic, however, lacking a…

Machine Learning · Computer Science 2026-01-08 Ye Su , Yong Liu

Mixture of Experts (MoE) are rising in popularity as a means to train extremely large-scale models, yet allowing for a reasonable computational cost at inference time. Recent state-of-the-art approaches usually assume a large number of…

Computer Vision and Pattern Recognition · Computer Science 2023-04-13 Amelie Royer , Ilia Karmanov , Andrii Skliar , Babak Ehteshami Bejnordi , Tijmen Blankevoort

Averaging the parameters of models that have the same architecture and initialization can provide a means of combining their respective capabilities. In this paper, we take the perspective that this "merging" operation can be seen as…

Machine Learning · Computer Science 2022-08-29 Michael Matena , Colin Raffel

The success of pretrain-finetune paradigm brings about the release of numerous model weights. In this case, merging models finetuned on different tasks to enable a single model with multi-task capabilities is gaining increasing attention…

Machine Learning · Computer Science 2024-09-30 Chenyu Huang , Peng Ye , Tao Chen , Tong He , Xiangyu Yue , Wanli Ouyang

Modern recommender systems trained on domain-specific data often struggle to generalize across multiple domains. Cross-domain sequential recommendation has emerged as a promising research direction to address this challenge; however,…

Information Retrieval · Computer Science 2026-01-06 Hyunsoo Kim , Jaewan Moon , Seongmin Park , Jongwuk Lee

Optimization in multi-task learning (MTL) is more challenging than single-task learning (STL), as the gradient from different tasks can be contradictory. When tasks are related, it can be beneficial to share some parameters among them…

Computer Vision and Pattern Recognition · Computer Science 2022-12-16 Zitian Chen , Yikang Shen , Mingyu Ding , Zhenfang Chen , Hengshuang Zhao , Erik Learned-Miller , Chuang Gan

The growing demand for large language models (LLMs) with tunable reasoning capabilities in many real-world applications highlights a critical need for methods that can efficiently produce a spectrum of models balancing reasoning depth and…

Artificial Intelligence · Computer Science 2025-09-30 Xiaochong Lan , Yu Zheng , Shiteng Cao , Yong Li

The recent rapid progress in (self) supervised learning models is in large part predicted by empirical scaling laws: a model's performance scales proportionally to its size. Analogous scaling laws remain elusive for reinforcement learning…

Large Language Models (LLMs) require instruction fine-tuning to perform different downstream tasks. However, the instruction fine-tuning phase still demands significant computational resources and labeled data, lacking a paradigm that can…

Computation and Language · Computer Science 2025-03-10 Yiguan Lin , Bin Xu , Yinghao Li , Yang Gao

There is a long history in machine learning of model ensembling, beginning with boosting and bagging and continuing to the present day. Much of this history has focused on combining models for classification and regression, but recently…

Machine Learning · Computer Science 2024-05-28 Ira Globus-Harris , Varun Gupta , Michael Kearns , Aaron Roth

Recent advances in large language models have led to specialized models excelling in specific domains, creating a need for efficient model merging techniques. While traditional merging approaches combine parameters into a single static…

Computation and Language · Computer Science 2025-05-26 Shuqi Liu , Yuxuan Yao , Bowei He , Zehua Liu , Xiongwei Han , Mingxuan Yuan , Han Wu , Linqi Song

Underlying data distributions of natural language, programming code, and mathematical symbols vary vastly, presenting a complex challenge for large language models (LLMs) that strive to achieve high performance across all three domains…

Computation and Language · Computer Science 2024-03-27 Ning Ding , Yulin Chen , Ganqu Cui , Xingtai Lv , Weilin Zhao , Ruobing Xie , Bowen Zhou , Zhiyuan Liu , Maosong Sun

Mixture-of-Experts (MoE) represents an ensemble methodology that amalgamates predictions from several specialized sub-models (referred to as experts). This fusion is accomplished through a router mechanism, dynamically assigning weights to…

Machine Learning · Computer Science 2024-03-27 Jinze Zhao , Peihao Wang , Zhangyang Wang

Training convolutional networks (CNN's) that fit on a single GPU with minibatch stochastic gradient descent has become effective in practice. However, there is still no effective method for training large CNN's that do not fit in the memory…

Computer Vision and Pattern Recognition · Computer Science 2017-04-24 Sam Gross , Marc'Aurelio Ranzato , Arthur Szlam

Transfer and multi-task learning have traditionally focused on either a single source-target pair or very few, similar tasks. Ideally, the linguistic levels of morphology, syntax and semantics would benefit each other by being trained in a…

Computation and Language · Computer Science 2017-07-25 Kazuma Hashimoto , Caiming Xiong , Yoshimasa Tsuruoka , Richard Socher

This paper addresses the limitations of a single agent in task decomposition and collaboration during complex task execution, and proposes a multi-agent architecture for modular task decomposition and dynamic collaboration based on large…

Artificial Intelligence · Computer Science 2025-11-04 Shuaidong Pan , Di Wu