No article text was recorded for this version.
zero-overhead weight merging
Version 2 of 2232 papers297 conceptsfrozen 2026-10-04
Papers in this version
- Blog: Chain of LoRA: Efficient Fine-tuning of Language Models via Residual Learning
- Blog: FLoRA: Fused forward-backward adapters for parameter efficient fine-tuning and reducing inference-time latencies of LLMs
- Blog: Zero-Shot Cross-Domain Dialogue State Tracking via Dual Low-Rank Adaptation
- Blog: AmoebaLLM: Constructing Any-Shape Large Language Models for Efficient and Instant Deployment
- Blog: SEAL: Scaling to Emphasize Attention for Long-Context Retrieval
- Blog: LoRA-Switch: Boosting the Efficiency of Dynamic LLM Adapters via System-Algorithm Co-design
- Blog: MoRA: Missing Modality Low-Rank Adaptation for Visual Recognition
- Blog: NeuroAda: Activating Each Neuron’s Potential for Parameter-Efficient Fine-Tuning
- Blog: Language and Task Arithmetic with Parameter-Efficient Layers for Zero-Shot Summarization
- Blog: HoRA: Cross-Head Low-Rank Adaptation with Joint Hypernetworks
- Blog: Steering Large Language Models using Conceptors: Improving Addition-Based Activation Engineering
- Blog: You Only Learn One Representation: Unified Network for Multiple Tasks
- Blog: Hydra: Multi-head Low-rank Adaptation for Parameter Efficient Fine-tuning
- Blog: Consolidator: Mergable Adapter with Group Connections for Visual Adaptation
- Blog: Modeling Associative Plasticity between Synapses to Enhance Learning of Spiking Neural Networks
- Blog: Orthogonal Projection Subspace to Aggregate Online Prior-knowledge for Continual Test-time Adaptation
- Blog: IntLoRA: Integral Low-rank Adaptation of Quantized Diffusion Models
- Blog: Parameter-Efficient Model Adaptation for Vision Transformers
- Blog: Regularized Mask Tuning: Uncovering Hidden Knowledge in Pre-trained Vision-Language Models
- Blog: Parameter-Efficient Fine-Tuning without Introducing New Latency
- Blog: Is Multiple Object Tracking a Matter of Specialization?
- Blog: AuroRA: Breaking Low-Rank Bottleneck of LoRA with Nonlinear Mapping
- Blog: NEAT: Nonlinear Parameter-efficient Adaptation of Pre-trained Models
- Blog: Parameter-efficient Fine-tuning for Vision Transformers
- Blog: LoRA: Low-Rank Adaptation of Large Language Models
- Blog: From Data Deluge to Data Curation: A Filtering-WoRA Paradigm for Efficient Text-based Person Search
- Blog: Efficient Adaptation of Large Vision Transformer via Adapter Re-Composing
- Blog: Scalable Weight Reparametrization for Efficient Transfer Learning
- Blog: Parameter-Efficient Fine-Tuning without Introducing New Latency
- Blog: Efficient Adaptation of Large Vision Transformer via Adapter Re-Composing
- Blog: Continuous Knowledge-Preserving Decomposition for Few-Shot Continual Learning
- Blog: Efficient Adaptation of Deep Neural Networks for Semantic Segmentation in Space Applications
- Blog: FacT: Factor-Tuning for Lightweight Adaptation on Vision Transformer
- Blog: Consolidator: Mergeable Adapter with Grouped Connections for Visual Adaptation
- Blog: KARST: Multi-Kernel Kronecker Adaptation with Re-Scaling Transmission for Visual Classification
- Blog: GRASP: GRouped Activation Shared Parameterization for Parameter-Efficient Fine-Tuning and Robust Inference of Transformers
- Blog: Efficient adaptation of deep neural networks for semantic segmentation in space applications
- Blog: Token Compensator: Altering Inference Cost of Vision Transformer Without Re-tuning
- Blog: Multimodal Pathway: Improve Transformers with Irrelevant Data from Other Modalities
- Blog: Efficient Fine-Tuning of Compressed Language Models with Learners
- Blog: SEAL: Scaling to Emphasize Attention for Long-Context Retrieval
- Blog: MoRA: Missing Modality Low-Rank Adaptation for Visual Recognition
- Blog: Long-tailed Recognition with Model Rebalancing
- Blog: LoRA-Gen: Specializing Large Language Model via Online LoRA Generation
- Blog: Learn to Preserve and Diversify: Parameter-Efficient Group with Orthogonal Regularization for Domain Generalization
- Blog: LoRA: Low-Rank Adaptation of Large Language Models
- Blog: Frosting Weights for Better Continual Training
- Blog: Low-Rank Adaptation for Foundation Models: A Comprehensive Review
- Blog: Align, Don't Divide: Revisiting the LoRA Architecture in Multi-Task Learning
- Blog: ROSA: Random Subspace Adaptation for Efficient Fine-Tuning
- Blog: Tensor Train Low-rank Approximation (TT-LoRA): Democratizing AI with Accelerated LLMs
- Blog: Decorrelation Speeds Up Vision Transformers
- Blog: LoRA-Gen: Specializing Large Language Model via Online LoRA Generation
- Blog: NAS-LoRA: Empowering Parameter-Efficient Fine-Tuning for Visual Foundation Models with Searchable Adaptation
- Blog: Parameter Efficient Quasi-Orthogonal Fine-Tuning via Givens Rotation
- Blog: Exploiting the Redundancy in Convolutional Filters for Parameter Reduction
- Blog: AdaptSR: Low-Rank Adaptation for Efficient and Scalable Real-World Super-Resolution
- Blog: PEAR: Position-Embedding-Agnostic Attention Re-weighting Enhances Retrieval-Augmented Generation with Zero Inference Overhead
- Blog: ViDA: Homeostatic Visual Domain Adapter for Continual Test Time Adaptation
- Blog: Efficient Learning of Multiple NLP Tasks via Collective Weight Factorization on BERT
Showing 60 of 232 papers.
History & citation
Current · Permanent link
Cite this direction
Latest version (a living page):
@misc{lacuna_zero_overhead_weight_merging_25e32235,
title = {zero-overhead weight merging},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/zero-overhead-weight-merging/rdr_25e322357cf944d4a081700ce585f9a6},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-22},
}This exact version (version 2, frozen text):
@misc{lacuna_zero_overhead_weight_merging_25e32235_v2,
title = {zero-overhead weight merging},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/zero-overhead-weight-merging/rdr_25e322357cf944d4a081700ce585f9a6/v/2},
note = {Version 2, frozen 2026-10-04. Accessed 2026-11-22},
}