MAny: Merge Anything for Multimodal Continual Instruction Tuning

• URL: http://arxiv.org/abs/2604.14016v1
• Date: Wed, 15 Apr 2026 15:57:23 GMT
• Title: MAny: Merge Anything for Multimodal Continual Instruction Tuning
• Abstract summary: textbfMAny (textbfMAny) is a framework that merges task-specific knowledge through textbfCross-modal textbfProjection textbfMerging.<n>textbfLow-rank textbfParameter textbfMerging (textbfLPM)
• Score: 52.50936513604062
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Multimodal Continual Instruction Tuning (MCIT) is essential for sequential task adaptation of Multimodal Large Language Models (MLLMs) but is severely restricted by catastrophic forgetting. While existing literature focuses on the reasoning language backbone, in this work, we expose a critical yet neglected dual-forgetting phenomenon across both perception drift in Cross-modal Projection Space and reasoning collapse in Low-rank Parameter Space. To resolve this, we present \textbf{MAny} (\textbf{M}erge \textbf{Any}thing), a framework that merges task-specific knowledge through \textbf{C}ross-modal \textbf{P}rojection \textbf{M}erging (\textbf{CPM}) and \textbf{L}ow-rank \textbf{P}arameter \textbf{M}erging (\textbf{LPM}). Specifically, CPM recovers perceptual alignment by adaptively merging cross-modal visual representations via visual-prototype guidance, ensuring accurate feature recovery during inference. Simultaneously, LPM eliminates mutual interference among task-specific low-rank modules by recursively merging low-rank weight matrices. By leveraging recursive least squares, LPM provides a closed-form solution that mathematically guarantees an optimal fusion trajectory for reasoning stability. Notably, MAny operates as a training-free paradigm that achieves knowledge merging via efficient CPU-based algebraic operations, eliminating additional gradient-based optimization beyond initial tuning. Our extensive evaluations confirm the superior performance and robustness of MAny across multiple MLLMs and benchmarks. Specifically, on the UCIT benchmark, MAny achieves significant leads of up to 8.57\% and 2.85\% in final average accuracy over state-of-the-art methods across two different MLLMs, respectively.

Related papers

• When Muon Meets Task Interference: A Spectral Perspective on Continual Learning and Model Merging [32.67767249149513]
  We show that the two difficulties are in fact two instances of the same phenomenon.<n>A parameter update useful for one task shifts the model's outputs on another.<n>We identify the recent Muon as a mechanism that regulates this factor by construction.
  arXiv  Detail & Related papers  (2026-08-27T10:56:58Z)
• Parameter-Efficient Modality-Balanced Symmetric Fusion for Multimodal Remote Sensing Semantic Segmentation [8.840077295284393]
  MoBaNet is a parameter-efficient and modality-balanced symmetric fusion framework.<n>Built upon a largely frozen VFM backbone, MoBaNet adopts a symmetric dual-stream architecture to preserve generalizable representations.<n>Experiments on the ISPRS Vaihingen and Potsdam benchmarks demonstrate that MoBaNet achieves state-of-the-art performance.
  arXiv  Detail & Related papers  (2026-03-18T13:23:58Z)
• Mitigating Modality Imbalance in Multi-modal Learning via Multi-objective Optimization [57.00656508727821]
  Multi-modal learning (MML) aims to integrate information from multiple modalities, which is expected to lead to superior performance over single-modality learning.<n>Recent studies have shown that MML can underperform, even compared to single-modality approaches, due to imbalanced learning across modalities.<n>We propose a gradient-based algorithm to solve the modified MML problem.
  arXiv  Detail & Related papers  (2025-11-10T04:16:01Z)
• Train with Perturbation, Infer after Merging: A Two-Stage Framework for Continual Learning [57.514786046966265]
  We propose textbfPerturb-and-Merge (P&M), a novel continual learning framework that integrates model merging into the CL paradigm to mitigate forgetting.<n>Our proposed approach achieves state-of-the-art performance on several continual learning benchmark datasets.
  arXiv  Detail & Related papers  (2025-05-28T14:14:19Z)
• Optimal Brain Iterative Merging: Mitigating Interference in LLM Merging [11.708743111945727]
  Large Language Models (LLMs) have demonstrated impressive capabilities, but their high computational costs pose challenges for customization.<n>Model merging offers a cost-effective alternative, yet existing methods suffer from interference among parameters, leading to performance degradation.<n>We propose Optimal Brain Iterative Merging, a novel method designed to mitigate both intra-model and inter-model interference.
  arXiv  Detail & Related papers  (2025-02-17T09:07:49Z)
• R-MTLLMF: Resilient Multi-Task Large Language Model Fusion at the Wireless Edge [78.26352952957909]
  Multi-task large language models (MTLLMs) are important for many applications at the wireless edge, where users demand specialized models to handle multiple tasks efficiently.<n>The concept of model fusion via task vectors has emerged as an efficient approach for combining fine-tuning parameters to produce an MTLLM.<n>In this paper, the problem of enabling edge users to collaboratively craft such MTLMs via tasks vectors is studied, under the assumption of worst-case adversarial attacks.
  arXiv  Detail & Related papers  (2024-11-27T10:57:06Z)
• Low-Rank Multitask Learning based on Tensorized SVMs and LSSVMs [65.42104819071444]
  Multitask learning (MTL) leverages task-relatedness to enhance performance. We employ high-order tensors, with each mode corresponding to a task index, to naturally represent tasks referenced by multiple indices. We propose a general framework of low-rank MTL methods with tensorized support vector machines (SVMs) and least square support vector machines (LSSVMs)
  arXiv  Detail & Related papers  (2023-08-30T14:28:26Z)
• Theoretical Convergence of Multi-Step Model-Agnostic Meta-Learning [63.64636047748605]
  We develop a new theoretical framework to provide convergence guarantee for the general multi-step MAML algorithm. In particular, our results suggest that an inner-stage step needs to be chosen inversely proportional to $N$ of inner-stage steps in order for $N$ MAML to have guaranteed convergence.
  arXiv  Detail & Related papers  (2020-02-18T19:17:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.