mlsanityData to 14 Oct 2026

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

arXiv 2511.03981 ↗First posted 6 Nov 2025cs.LGcited by 4

Not on the map. Why?

Abstract (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.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 11 of its 25 references that are in the map. First posted on arXiv 6 Nov 2025.

Builds on

11 of 25 references.

References · 11

  1. 2021Combinatorial optimization and reasoning with graph neural networkscited by 268
  2. 2021End-to-End Constrained Optimization Learning: A Surveycited by 157
  3. 2023MolCA: Molecular Graph-Language Modeling with Cross-Modal Projector and Uni-Modal Adaptercited by 128
  4. 2023Modular Deep Learningcited by 83
  5. 2024GraphEdit: Large Language Models for Graph Structure Learningcited by 33
  6. 2022GPS++: An Optimised Hybrid MPNN/Transformer for Molecular Property Predictioncited by 28
  7. 2025Dynamic Prompt Fusion for Multi-Task and Cross-Domain Adaptation in LLMscited by 9
  8. 2025Contrastive Learning-Based Dependency Modeling for Anomaly Detection in Cloud Servicescited by 9
  9. 2025Integrating Structure-Aware Attention and Knowledge Graphs in Explainable Recommendation Systemscited by 4
  10. 2020An Experimental Study of the Transferability of Spectral Graph Networkscited by 3
  11. 2025Controllable Abstraction in Summary Generation for Large Language Models via Prompt Engineeringcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 1.

4 papers.

  1. 2025Parameter-Efficient Fine-Tuning with Differential Privacy for Robust Instruction Adaptation in Large Language Modelscited by 5
  2. 2025Deep Q-Learning-Based Intelligent Scheduling for ETL Optimization in Heterogeneous Data Environmentscited by 4
  3. 2025Advancing Text Classification with Large Language Models and Neural Attention Mechanismscited by 1
  4. 2026Coordinated Semantic Alignment and Evidence Constraints for Retrieval-Augmented Generation with Large Language Modelscited by 0