paper-with-me

홈 › Papers

Structural Priors and Modular Adapters in the Composable Fine-Tuning Algorithm of Large-Scale Models

2025-11-06 · Yuxiao Wang, Di Wu, Feng Liu, Zhimin Qiu, Chenrui Hu arxiv

This paper proposes a composable fine-tuning method that integrates graph structural priors with modular adapters to address the high computational cost and structural instability faced by large-scale pre-trained models in multi-task adaptation. The method introduces a relation matrix to model dependencies among tasks, explicitly encoding correlations between nodes and paths into graph structural priors, which provide unified structural constraints for adapter weight allocation and path selection. Modular adapters are embedded into different layers through low-rank mapping and a pluggable mechanism, enabling efficient cross-task composition and reuse under prior guidance. This mechanism not only improves parameter efficiency and training stability but also alleviates path conflicts and redundant computation in multi-task scenarios. Furthermore, experiments on hyperparameter sensitivity, environmental sensitivity, and data sensitivity are conducted to systematically analyze key factors such as routing temperature, gating thresholds, and relation matrix regularization strength, verifying the consistency and superior performance of the method under structural constraints. The results demonstrate that the proposed framework significantly enhances task prediction accuracy, adapter weight allocation precision, and overall computational efficiency while maintaining model lightweight design, highlighting the synergistic advantages of graph priors and modular mechanisms in composable fine-tuning.

📄 PDF Abstract BibTeX arXiv:2511.03981

Code (0)

등록된 구현이 없습니다.

Tasks

Computational Efficiency

Similar Papers 제목 키워드 기반

Composable Sparse Fine-Tuning for Cross-Lingual Transfer

2021-10-14 · ACL 2022 5 · Alan Ansell, Edoardo Maria Ponti, Anna Korhonen, Ivan Vulić

Fine-tuning the entire set of parameters of a large pretrained model has become the mainstream approach for transfer learning. To increase its efficiency and prevent catastrophic forgetting and interference, techniques l…

Cross-Lingual TransferLanguage ModelingLanguage ModellingMasked Language Modeling+2

MoLoRA: Composable Specialization via Per-Token Adapter Routing

2026-03-16 · Shrey Shah, Justin Wagle arxiv

Multi-adapter serving systems route entire sequences to a single adapter, forcing a choice when requests span multiple domains. This assumption fails in two important settings: (1) multimodal generation, where text and i…

multimodal generation

CharCom: Composable Identity Control for Multi-Character Story Illustration

2025-10-11 · Zhongsheng Wang, Ming Lin, Zhedong Lin, Yaser Shakib 외 arxiv

Ensuring character identity consistency across varying prompts remains a fundamental limitation in diffusion-based text-to-image generation. We propose CharCom, a modular and parameter-efficient framework that achieves c…

Text-to-Image Generation

Composable Sparse Fine-Tuning for Cross-Lingual Transfer

2021-11-16 · ACL ARR November 2021 11 · Anonymous

Fine-tuning the entire set of parameters of a large pretrained model has become the mainstream approach for transfer learning. To increase its efficiency and prevent catastrophic forgetting and interference, techniques l…

Cross-Lingual TransferLanguage ModelingLanguage ModellingMasked Language Modeling+2

FMTK: A Modular Toolkit for Composable Time Series Foundation Model Pipelines

2025-11-30 · Hetvi Shastri, Pragya Sharma, Walid A. Hanafy, Mani Srivastava 외 arxiv

Foundation models (FMs) have opened new avenues for machine learning applications due to their ability to adapt to new and unseen tasks with minimal or no further training. Time-series foundation models (TSFMs) -- FMs tr…