mlsanityData to 7 Oct 2026

Cross-Modal Projection in Multimodal LLMs Doesn't Really Project Visual Attributes to Textual Space

arXiv 2402.16832 ↗First posted 26 Feb 2024cs.CL cs.AI cs.CVcited by 21

On the map: Multimodal LLMs / vision-language models → MLLM training and visual instruction tuning

Abstract (arXiv)

Multimodal large language models (MLLMs) like LLaVA and GPT-4(V) enable general-purpose conversations about images with the language modality. As off-the-shelf MLLMs may have limited capabilities on images from domains like dermatology and agriculture, they must be fine-tuned to unlock domain-specific applications. The prevalent architecture of current open-source MLLMs comprises two major modules: an image-language (cross-modal) projection network and a large language model. It is desirable to understand the roles of these two modules in modeling domain-specific visual attributes to inform the design of future models and streamline the interpretability efforts on the current models. To this end, via experiments on 4 datasets and under 2 fine-tuning settings, we find that as the MLLM is fine-tuned, it indeed gains domain-specific visual capabilities, but the updates do not lead to the projection extracting relevant domain-specific visual attributes. Our results indicate that the domain-specific visual attributes are modeled by the LLM, even when only the projection is fine-tuned. Through this study, we offer a potential reinterpretation of the role of cross-modal projections in MLLM architectures. Project webpage: https://claws-lab.github.io/projection-in-MLLMs/

Key facts. Cited by 21 papers in the map (data through 7 Oct 2026). Placed in MLLM training and visual instruction tuning, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 18 of its 35 references that are in the map. First posted on arXiv 26 Feb 2024.

Builds on

18 of 35 references.

Closest ancestors · 3

Same direction, newest first

  1. 2023Finding and Editing Multi-Modal Neurons in Pre-Trained Transformers13 Nov
  2. 2023Improved Baselines with Visual Instruction Tuning5 Oct
  3. 2023InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning11 May

Other references · 15

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2023GPT-4 Technical Reportcited by 18,132
  4. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  5. 2019HuggingFace's Transformers: State-of-the-art Natural Language Processingcited by 2,201
  6. 2013Describing Textures in the Wildcited by 2,110
  7. 2023Video-LLaVA: Learning United Visual Representation by Alignment Before Projectioncited by 1,385
  8. 2023LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Daycited by 1,287
  9. 2020Concept Bottleneck Modelscited by 1,022
  10. 2019Learning Representations by Maximizing Mutual Information Across Viewscited by 733
  11. 2023"Kelly is a Warm Person, Joseph is a Role Model": Gender Biases in LLM-Generated Reference Letterscited by 196
  12. 2023AnyMAL: An Efficient and Scalable Any-Modality Augmented Language Modelcited by 93
3 more
  1. 2018CrisisMMD: Multimodal Twitter Datasets from Natural Disasterscited by 88
  2. 2023Multimodal Neurons in Pretrained Text-Only Transformerscited by 49
  3. 2020Analysis of Social Media Data using Multimodal Deep Learning for Disaster Responsecited by 23

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 5; 2025: 11; 2026: 5.

21 papers.

Sort:

By area:

  1. 2024CALF: Aligning LLMs for Time Series Forecasting via Cross-modal Fine-Tuningcited by 105
  2. 2024Explainable and Interpretable Multimodal Large Language Models: A Comprehensive Surveycited by 52
  3. 2025Visual Representation Alignment for Multimodal Large Language Modelscited by 40
  4. 2025Visual Large Language Models for Generalized and Specialized Applicationscited by 35
  5. 2025Learning to See Before Seeing: Demystifying LLM Visual Priors from Language Pre-trainingcited by 23
  6. 2024LUMIA: Linear probing for Unimodal and MultiModal Membership Inference Attacks leveraging internal LLM statescited by 8
  7. 2025Turning Internal Gap into Self-Improvement: Promoting the Generation-Understanding Unification in MLLMscited by 8
  8. 2025Merge then Realign: Simple and Effective Modality-Incremental Continual Learning for Multimodal LLMscited by 6
  9. 2026Do MLLMs Really See It: Reinforcing Visual Attention in Multimodal LLMscited by 4
  10. 2025LLMs Can Compensate for Deficiencies in Visual Representationscited by 3
  11. 2025MonoSR: Open-Vocabulary Spatial Reasoning from Monocular Imagescited by 3
  12. 2024Phase Diagram of Vision Large Language Models Inference: A Perspective from Interaction across Image and Instructioncited by 2
9 more
  1. 2025A Framework for Situating Innovations, Opportunities, and Challenges in Advancing Vertical Systems with Large AI Modelscited by 2
  2. 2025A Survey on Interpretability in Visual Recognitioncited by 2
  3. 2024Intriguing Properties of Large Language and Vision Modelscited by 1
  4. 2026Lost in the Hype: Revealing and Dissecting the Performance Degradation of Medical Multimodal Large Language Models in Image Classificationcited by 1
  5. 2026CheXmix: Unified Generative Pretraining for Vision Language Models in Medical Imagingcited by 1
  6. 2025CircuitProbe: Tracing Visual Temporal Evidence Flow in Video Language Modelscited by 0
  7. 2025LLM-based Fusion of Multi-modal Features for Commercial Memorability Predictioncited by 0
  8. 2026SSDA: Bridging Spectral and Structural Gaps via Dual Adaptation for Vision-Based Time Series Forecastingcited by 0
  9. 2026Abstracting Cross-Domain Action Sequences into Interpretable Workflowscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023How to Bridge the Gap between Modalities: Survey on Multimodal Large Language Model
  2. 2025Same Content, Different Answers: Cross-Modal Inconsistency in MLLMs
  3. 2024The Revolution of Multimodal Large Language Models: A Survey
  4. 2025Multi-Layer Visual Feature Fusion in Multimodal LLMs: Methods, Analysis, and Best Practices
  5. 2024A Comprehensive Survey and Guide to Multimodal Large Language Models in Vision-Language Tasks
  6. 2024CoF: Coarse to Fine-Grained Image Understanding for Multi-modal Large Language Models
  7. 2024Why do LLaVA Vision-Language Models Reply to Images in English?
  8. 2026TokenSwap: Benchmarking and Reducing the Modality Gap in Multimodal LLMs