Parameter Sharing

Follow this topic Ask about this topic

Latest papers

12 papers3 papers in the last four weeks, with none the four weeks before. 0.0% of all new papers.

Compose the Horizon: Parameter-Shared Blockwise Forecasting under a Sequential-Compute Trade-off
Oct 17, 2026Zeyan Li, Weiheng Li, Shengda Zhuo +3Long-Term Time Series Forecasting Time Series Forecasting
Does Every User Need a Private LoRA? Decoupling Personalization from Per-User Adaptation
Oct 1, 2026Songyuan Sui, Srikanth Malla, Chiho Choi +1LLM Personalization Personalized Language Models
Samsung Semiconductor, US · Rice University
Statistical Gains from Looped Estimation under Parameter Budgets
Sep 22, 2026Xinyu Tian, Xiaotong ShenParameter Estimation Parameter Sharing
School of Statistics, University of Minnesota
SAPE: Sandwich Adapters for Parameter Efficiency in Large Language Model Fine-Tuning
Aug 15, 2026Mohammad Aref Jafari-Raddani, Morteza Mohajjel KafshdoozParameter Sharing Parameter-Efficient Fine-Tuning
Department of Electrical and Computer Engineering, Qom University of Technology, Qom, Iran.
FEAST: Federated Shared-Space Training for Resource-Heterogeneous Clients
Aug 10, 2026Bostan Khan, Masoud DaneshtalabDynamic Neural Networks Communication-Efficient Distributed Training
Mälardalen University, Västerås, Sweden
ParVL: Parallel Scaling and Expandable Compute Allocation for Multimodal LLMs
Aug 4, 2026Yang Yang, Qinyu Zhao, Mouxiang Chen +5Vision-Language Models Multimodal Large Language Models
1The Australian National University · 2Shanghai AI Laboratory · 3Zhejiang University +2
Dualformer: Efficient Feature Extractor for Complex-valued Blind Communication Signal Analysis
Jun 30, 2026Yurui Zhao, Xiang Wang, Jingreng Lei +3Transformer Automatic Modulation Classification
College of Electronic Science and Technology, National University of Defense Technology, Changsha 410073, China. · Department of Electrical and Computer Engineering, The University of Hong Kong
FedSPC: Shared Parameter Correction for Personalized Federated Learning
Jun 11, 2026Kannanthodath Induchoodan Ajay Menon, Christian Prehofer, Yunfei Xu +1Non-IID Federated Learning Federated Learning
DENSO AUTOMOTIVE Deutschland GmbH, Germany · Technical University of Munich, Germany · DENSO International America, Inc., United States
Multi-Mixer Models: Flexible Sequence Modeling with Shared Representations
May 27, 2026Kevin Y. Li, Asher Trockman, Ananda Theertha Suresh +1Parameter Sharing Transformer Attention
Carnegie Mellon University · Google Research
Reshape and Recur: Improving SSMs with Input Reshaping and Depth Recurrence
May 15, 2026Mónika Farsang, Ramin Hasani, Daniela Rus +1Parameter Sharing State Space Models
1TU Wien · MIT CSAIL · Liquid AI
Light-ResKAN: A Parameter-Sharing Lightweight KAN with Gram Polynomials for Efficient SAR Image Recognition
Apr 2, 2026Pan Yi, Weijie Li, Xiaodong Chen +3Parameter Sharing Kolmogorov-Arnold Networks
The authors are with the College of Electronic Science and Technology, National University of Defense Technology, Changsha, 410073, China.
Automated Gradient-Driven Parameter Sharing for Low-Resource Multilingual Speech-to-Text Translation
Mar 26, 2026Ruiyan Sun, Satoshi NakamuraSpeech Translation Parameter Sharing
School of Data Science · School of Aritificial Intelligence, The Chinese University of Hong Kong,Shenzhen
SRNeRV: A Scale-wise Recursive Framework for Neural Video Representation
Mar 9, 2026Jia Wang, Jun Zhu, Xinfeng ZhangParameter Sharing Implicit Neural Representations
School of Computer Science and Technology, University of Chinese Academy of Sciences, Beijing, China

Oct 17, 2026J/K move · Enter open · S save

Compose the Horizon: Parameter-Shared Blockwise Forecasting under a Sequential-Compute Trade-off

Zeyan Li, Weiheng Li, Shengda Zhuo, Libing Chen+2

Long-Term Time Series Forecasting Time Series Forecasting Multivariate Time Series Forecasting Parameter Sharing

Open paper AskPDF arXiv

Most long-horizon forecasters emit the complete future in one pass. We ask whether a compact operator can instead be reused to emit contiguous future blocks, trading stored parameters for sequential computation. Recursive Functional Decomposition (RFD) factorizes the horizon mapping into three parameter-shared block predictions. The blocks occupy disjoint future intervals and are concatenated, not added as trend, seasonal, or residual components, and each emitted block joins the rolling context that conditions the next application. In controlled three-seed experiments, removing forward feedback hurts all four evaluated datasets while stopping gradients through it changes little, so the useful signal is the evolving forecast state itself; reapplying the shared operator helps consistently, but by less than widening it. The model uses 0.08--0.46M parameters, yet its sequential channel-wise execution is slower and more memory-intensive than an iterative baseline at high channel count. In a frozen long-context evaluation against ten strong baseline families, RFD records lower test error in 23 of 24 cells; the comparison is cross-protocol, with RFD using 512/720-step context and the baselines using 96.