Sample-efficient Integration of New Modalities into Large Language Models

• URL: http://arxiv.org/abs/2509.04606v1
• Date: Thu, 04 Sep 2025 18:41:59 GMT
• Title: Sample-efficient Integration of New Modalities into Large Language Models
• Authors: Osman Batur İnce, André F. T. Martins, Oisin Mac Aodha, Edoardo M. Ponti,
• Abstract summary: Multimodal foundation models can process several modalities.<n>We introduce a method for sample-efficient modality integration into Large Language Models.<n>We find that SEMI achieves a significant boost in sample efficiency during few-shot integration of new modalities.
• Score: 48.81776019848246
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multimodal foundation models can process several modalities. However, since the space of possible modalities is large and evolving over time, training a model from scratch to encompass all modalities is unfeasible. Moreover, integrating a modality into a pre-existing foundation model currently requires a significant amount of paired data, which is often not available for low-resource modalities. In this paper, we introduce a method for sample-efficient modality integration (SEMI) into Large Language Models (LLMs). To this end, we devise a hypernetwork that can adapt a shared projector -- placed between modality-specific encoders and an LLM -- to any modality. The hypernetwork, trained on high-resource modalities (i.e., text, speech, audio, video), is conditioned on a few samples from any arbitrary modality at inference time to generate a suitable adapter. To increase the diversity of training modalities, we artificially multiply the number of encoders through isometric transformations. We find that SEMI achieves a significant boost in sample efficiency during few-shot integration of new modalities (i.e., satellite images, astronomical images, inertial measurements, and molecules) with encoders of arbitrary embedding dimensionality. For instance, to reach the same accuracy as 32-shot SEMI, training the projector from scratch needs 64$\times$ more data. As a result, SEMI holds promise to extend the modality coverage of foundation models.

Related papers

• TokaMind: A Multi-Modal Transformer Foundation Model for Tokamak Plasma Dynamics [56.073642366268764]
  TokaMind is an open-source foundation model framework for fusion plasma modeling.<n>It is trained on heterogeneous tokamak diagnostics from the publicly available MAST dataset.<n>We evaluate TokaMind on the recently introduced MAST benchmark TokaMark.
  arXiv  Detail & Related papers  (2026-02-16T12:26:07Z)
• NExT-OMNI: Towards Any-to-Any Omnimodal Foundation Models with Discrete Flow Matching [64.10695425442164]
  We introduce NExT-OMNI, an open-source omnimodal foundation model that achieves unified modeling through discrete flow paradigms.<n>Trained on large-scale interleaved text, image, video, and audio data, NExT-OMNI delivers competitive performance on multimodal generation and understanding benchmarks.<n>To advance further research, we release training details, data protocols, and open-source both the code and model checkpoints.
  arXiv  Detail & Related papers  (2025-10-15T16:25:18Z)
• Unifying Multimodal Large Language Model Capabilities and Modalities via Model Merging [103.98582374569789]
  Model merging aims to combine multiple expert models into a single model, thereby reducing storage and serving costs.<n>Previous studies have primarily focused on merging visual classification models or Large Language Models (LLMs) for code and math tasks.<n>We introduce the model merging benchmark for MLLMs, which includes multiple tasks such as VQA, Geometry, Chart, OCR, and Grounding, providing both LoRA and full fine-tuning models.
  arXiv  Detail & Related papers  (2025-05-26T12:23:14Z)
• Multimodal Lego: Model Merging and Fine-Tuning Across Topologies and Modalities in Biomedicine [10.774128925670183]
  Multimodal Lego (MM-Lego) is a general-purpose fusion framework to turn any set of encoders into a competitive multimodal model with no or minimal fine-tuning.<n>We show that MM-Lego can be used as a model merging method which achieves competitive performance with end-to-end fusion models without any fine-tuning.
  arXiv  Detail & Related papers  (2024-05-30T11:14:01Z)
• Model Composition for Multimodal Large Language Models [71.5729418523411]
  We propose a new paradigm through the model composition of existing MLLMs to create a new model that retains the modal understanding capabilities of each original model. Our basic implementation, NaiveMC, demonstrates the effectiveness of this paradigm by reusing modality encoders and merging LLM parameters.
  arXiv  Detail & Related papers  (2024-02-20T06:38:10Z)
• Toward Robust Multimodal Learning using Multimodal Foundational Models [30.755818450393637]
  We propose TRML, Toward Robust Multimodal Learning using Multimodal Foundational Models. TRML employs generated virtual modalities to replace missing modalities. We also design a semantic matching learning module to align semantic spaces generated and missing modalities.
  arXiv  Detail & Related papers  (2024-01-20T04:46:43Z)
• What Makes for Robust Multi-Modal Models in the Face of Missing Modalities? [35.19295402483624]
  We model the scenarios of multi-modal models encountering missing modalities from an information-theoretic perspective. We introduce Uni-Modal Ensemble with Missing Modality Adaptation (UME-MMA) UME-MMA employs uni-modal pre-trained weights for the multi-modal model to enhance feature extraction and utilizes missing modality data augmentation techniques to better adapt to situations with missing modalities.
  arXiv  Detail & Related papers  (2023-10-10T07:47:57Z)
• eP-ALM: Efficient Perceptual Augmentation of Language Models [70.47962271121389]
  We propose to direct effort to efficient adaptations of existing models, and propose to augment Language Models with perception. Existing approaches for adapting pretrained models for vision-language tasks still rely on several key components that hinder their efficiency. We show that by freezing more than 99% of total parameters, training only one linear projection layer, and prepending only one trainable token, our approach (dubbed eP-ALM) significantly outperforms other baselines on VQA and Captioning.
  arXiv  Detail & Related papers  (2023-03-20T19:20:34Z)
• High-Modality Multimodal Transformer: Quantifying Modality & Interaction Heterogeneity for High-Modality Representation Learning [112.51498431119616]
  This paper studies efficient representation learning for high-modality scenarios involving a large set of diverse modalities. A single model, HighMMT, scales up to 10 modalities (text, image, audio, video, sensors, proprioception, speech, time-series, sets, and tables) and 15 tasks from 5 research areas.
  arXiv  Detail & Related papers  (2022-03-02T18:56:20Z)

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.