Multi-Layer Visual Feature Fusion in Multimodal LLMs: Methods, Analysis, and Best Practices

• URL: http://arxiv.org/abs/2503.06063v1
• Date: Sat, 08 Mar 2025 05:10:55 GMT
• Title: Multi-Layer Visual Feature Fusion in Multimodal LLMs: Methods, Analysis, and Best Practices
• Authors: Junyan Lin, Haoran Chen, Yue Fan, Yingqi Fan, Xin Jin, Hui Su, Jinlan Fu, Xiaoyu Shen,
• Abstract summary: Multimodal Large Language Models (MLLMs) have made significant advancements in recent years, with visual features playing an increasingly critical role in enhancing model performance.<n>However, the integration of multi-layer visual features in MLLMs remains underexplored, particularly with regard to optimal layer selection and fusion strategies.<n>This paper systematically investigates two core aspects of multi-layer visual feature fusion: (1) selecting the most effective visual layers and (2) identifying the best fusion approach with the language model.
• Score: 40.48590954153895
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multimodal Large Language Models (MLLMs) have made significant advancements in recent years, with visual features playing an increasingly critical role in enhancing model performance. However, the integration of multi-layer visual features in MLLMs remains underexplored, particularly with regard to optimal layer selection and fusion strategies. Existing methods often rely on arbitrary design choices, leading to suboptimal outcomes. In this paper, we systematically investigate two core aspects of multi-layer visual feature fusion: (1) selecting the most effective visual layers and (2) identifying the best fusion approach with the language model. Our experiments reveal that while combining visual features from multiple stages improves generalization, incorporating additional features from the same stage typically leads to diminished performance. Furthermore, we find that direct fusion of multi-layer visual features at the input stage consistently yields superior and more stable performance across various configurations. We make all our code publicly available: https://github.com/EIT-NLP/Layer_Select_Fuse_for_MLLM.

Related papers

• Rethinking Visual Layer Selection in Multimodal LLMs [46.091556112958884]
  This work proposes a Layer-wise Similarity approach to group CLIP-ViT layers with similar behaviors into shallow, middle, and deep categories. We revisit the visual layer selection problem in MLLMs at scale, training LLaVA-style models ranging from 1.4B to 7B parameters. We find that: (1) deep layers are essential for OCR tasks; (2) shallow and middle layers substantially outperform deep layers on reasoning tasks involving counting, positioning, and object localization; and (3) a lightweight fusion of features across shallow, middle, and deep layers consistently outperforms specialized fusion baselines and single-
  arXiv  Detail & Related papers  (2025-04-30T09:07:10Z)
• TAMP: Token-Adaptive Layerwise Pruning in Multimodal Large Language Models [23.916205754112774]
  Multimodal Large Language Models (MLLMs) have shown remarkable versatility in understanding diverse multimodal data and tasks. We propose TAMP, a simple yet effective pruning framework tailored for MLLMs. We validate our method on two state-of-the-art MLLMs: LLaVA-NeXT, designed for vision-language tasks, and VideoLLaMA2, capable of processing audio, visual, and language modalities.
  arXiv  Detail & Related papers  (2025-04-14T05:44:38Z)
• Layer-Aware Embedding Fusion for LLMs in Text Classifications [1.4250487522292254]
  We propose a layer-aware embedding selection method and investigate how to quantitatively evaluate different layers to identify the most important ones for downstream NLP tasks. Experiments on four English text classification datasets demonstrate that different layers in LLMs exhibit varying degrees of representational strength for classification. We also explore how combining embeddings from multiple LLMs, without requiring model fine-tuning, can improve performance.
  arXiv  Detail & Related papers  (2025-04-08T07:45:50Z)
• LayAlign: Enhancing Multilingual Reasoning in Large Language Models via Layer-Wise Adaptive Fusion and Alignment Strategy [33.85811169010525]
  Large language models (LLMs) exhibit suboptimal performance on low-resource languages.<n>Recent approaches have leveraged multilingual encoders alongside LLMs by introducing trainable parameters connecting the two models.<n>We propose aname, a framework that integrates representations from all encoder layers.
  arXiv  Detail & Related papers  (2025-02-17T03:45:03Z)
• Instruction-Guided Fusion of Multi-Layer Visual Features in Large Vision-Language Models [50.98559225639266]
  We investigate the contributions of visual features from different encoder layers using 18 benchmarks spanning 6 task categories.<n>Our findings reveal that multilayer features provide complementary strengths with varying task dependencies, and uniform fusion leads to suboptimal performance.<n>We propose the instruction-guided vision aggregator, a module that dynamically integrates multi-layer visual features based on textual instructions.
  arXiv  Detail & Related papers  (2024-12-26T05:41:31Z)
• EMMA: Efficient Visual Alignment in Multi-Modal LLMs [56.03417732498859]
  EMMA is a lightweight cross-modality module designed to efficiently fuse visual and textual encodings. EMMA boosts performance across multiple tasks by up to 9.3% while significantly improving robustness against hallucinations.
  arXiv  Detail & Related papers  (2024-10-02T23:00:31Z)
• NoteLLM-2: Multimodal Large Representation Models for Recommendation [71.87790090964734]
  Large Language Models (LLMs) have demonstrated exceptional proficiency in text understanding and embedding tasks. Their potential in multimodal representation, particularly for item-to-item (I2I) recommendations, remains underexplored. We propose an end-to-end fine-tuning method that customizes the integration of any existing LLMs and vision encoders for efficient multimodal representation.
  arXiv  Detail & Related papers  (2024-05-27T03:24:01Z)
• U3M: Unbiased Multiscale Modal Fusion Model for Multimodal Semantic Segmentation [63.31007867379312]
  We introduce U3M: An Unbiased Multiscale Modal Fusion Model for Multimodal Semantics. We employ feature fusion at multiple scales to ensure the effective extraction and integration of both global and local features. Experimental results demonstrate that our approach achieves superior performance across multiple datasets.
  arXiv  Detail & Related papers  (2024-05-24T08:58:48Z)
• Browse and Concentrate: Comprehending Multimodal Content via prior-LLM Context Fusion [70.9767518332692]
  Multimodal Large Language Models (MLLMs) that incorporate LLMs with pre-trained vision models have recently demonstrated impressive performance across diverse vision-language tasks. However, they fall short to comprehend context involving multiple images. We propose a two phase paradigm, browse-and-concentrate, to enable in-depth multimodal context fusion.
  arXiv  Detail & Related papers  (2024-02-19T14:59:07Z)
• From CLIP to DINO: Visual Encoders Shout in Multi-modal Large Language Models [36.41816380074965]
  We investigate the effectiveness of different vision encoders within Large Language Models (MLLMs) Our findings reveal that the shallow layer features of CLIP offer particular advantages for fine-grained tasks such as grounding and region understanding. We propose a simple yet effective feature merging strategy, named COMM, that integrates CLIP and DINO with Multi-level features Merging.
  arXiv  Detail & Related papers  (2023-10-13T02:41:55Z)

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.