Cross-Modal Knowledge Distillation without Paired Data: Theoretical Foundation and Algorithm

• URL: http://arxiv.org/abs/2606.10504v1
• Date: Tue, 09 Jun 2026 07:29:12 GMT
• Title: Cross-Modal Knowledge Distillation without Paired Data: Theoretical Foundation and Algorithm
• Abstract summary: Cross-modal knowledge distillation studies how a teacher model trained on one type of data can guide a student model building on another type of data.<n>Existing CMKD methods often require paired multi-modal data with aligned semantics.<n>We develop a new CMKD framework for the more challenging setting where paired data are unavailable.
• Score: 15.156476515650693
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Cross-modal knowledge distillation (CMKD) studies how a (large) teacher model trained on one type of data (e.g., images) can guide a (smaller) student model building on another type of data (e.g., text/audio). Existing CMKD methods often require paired multi-modal data with aligned semantics, but obtaining such paired data are often costly and impractical. To mitigate this limitation, we develop a new CMKD framework for the more challenging setting where paired data are unavailable. In particular, we establish a cross-modal distributional relationship between teacher and student models, which reveals two fundamental quantities governing effective distillation: feature alignment and label alignment. These quantities characterize semantic discrepancy between modalities at the levels of representation and prediction distributions, respectively. Motivated by this insight, we propose a principled framework, with theoretical guarantees, that enables effective cross-modal knowledge distillation by aligning distributions rather than individual samples. Extensive experiments across a wide range of multimodal benchmarks show that our framework is highly effective in both unpaired and paired data settings, improving significantly over prior work.

Related papers

• Multi-Modal Dataset Distillation in the Wild [75.64263877043615]
  We propose Multi-modal dataset Distillation in the Wild, i.e., MDW, to distill noisy multi-modal datasets into compact clean ones for effective and efficient model training.<n>Specifically, MDW introduces learnable fine-grained correspondences during distillation and adaptively optimize distilled data to emphasize correspondence-discriminative regions.<n>Extensive experiments validate MDW's theoretical and empirical efficacy with remarkable scalability, surpassing prior methods by over 15% across various compression ratios.
  arXiv  Detail & Related papers  (2025-06-02T12:18:20Z)
• Enhancing Knowledge Distillation of Large Language Models through Efficient Multi-Modal Distribution Alignment [10.104085497265004]
  We propose Ranking Loss based Knowledge Distillation (RLKD), which encourages consistency of peak predictions between the teacher and student models.<n>Our method enables the student model to better learn the multi-modal distributions of the teacher model, leading to a significant performance improvement in various downstream tasks.
  arXiv  Detail & Related papers  (2024-09-19T08:06:42Z)
• Multi Teacher Privileged Knowledge Distillation for Multimodal Expression Recognition [58.41784639847413]
  Human emotion is a complex phenomenon conveyed and perceived through facial expressions, vocal tones, body language, and physiological signals. In this paper, a multi-teacher PKD (MT-PKDOT) method with self-distillation is introduced to align diverse teacher representations before distilling them to the student. Results indicate that our proposed method can outperform SOTA PKD methods.
  arXiv  Detail & Related papers  (2024-08-16T22:11:01Z)
• DisCoM-KD: Cross-Modal Knowledge Distillation via Disentanglement Representation and Adversarial Learning [3.763772992906958]
  Cross-modal knowledge distillation (CMKD) refers to the scenario in which a learning framework must handle training and test data that exhibit a modality mismatch. DisCoM-KD (Disentanglement-learning based Cross-Modal Knowledge Distillation) explicitly models different types of per-modality information.
  arXiv  Detail & Related papers  (2024-08-05T13:44:15Z)
• A Generalization Theory of Cross-Modality Distillation with Contrastive Learning [49.35244441141323]
  Cross-modality distillation arises as an important topic for data modalities containing limited knowledge. We formulate a general framework of cross-modality contrastive distillation (CMCD), built upon contrastive learning. Our algorithm outperforms existing algorithms consistently by a margin of 2-3% across diverse modalities and tasks.
  arXiv  Detail & Related papers  (2024-05-06T11:05:13Z)
• Distilling Privileged Multimodal Information for Expression Recognition using Optimal Transport [46.91791643660991]
  Deep learning models for multimodal expression recognition have reached remarkable performance in controlled laboratory environments. These models struggle in the wild because of the unavailability and quality of modalities used for training. In practice, only a subset of the training-time modalities may be available at test time. Learning with privileged information enables models to exploit data from additional modalities that are only available during training.
  arXiv  Detail & Related papers  (2024-01-27T19:44:15Z)
• Multimodal Learning Without Labeled Multimodal Data: Guarantees and Applications [90.6849884683226]
  We study the challenge of interaction quantification in a semi-supervised setting with only labeled unimodal data. Using a precise information-theoretic definition of interactions, our key contribution is the derivation of lower and upper bounds. We show how these theoretical results can be used to estimate multimodal model performance, guide data collection, and select appropriate multimodal models for various tasks.
  arXiv  Detail & Related papers  (2023-06-07T15:44:53Z)
• CMD: Self-supervised 3D Action Representation Learning with Cross-modal Mutual Distillation [130.08432609780374]
  In 3D action recognition, there exists rich complementary information between skeleton modalities. We propose a new Cross-modal Mutual Distillation (CMD) framework with the following designs. Our approach outperforms existing self-supervised methods and sets a series of new records.
  arXiv  Detail & Related papers  (2022-08-26T06:06:09Z)
• Modality-specific Distillation [30.190082262375395]
  We propose modality-specific distillation (MSD) to effectively transfer knowledge from a teacher on multimodal datasets. Our idea aims at mimicking a teacher's modality-specific predictions by introducing an auxiliary loss term for each modality. Because each modality has different importance for predictions, we also propose weighting approaches for the auxiliary losses.
  arXiv  Detail & Related papers  (2021-01-06T05:45:07Z)
• MixKD: Towards Efficient Distillation of Large-scale Language Models [129.73786264834894]
  We propose MixKD, a data-agnostic distillation framework, to endow the resulting model with stronger generalization ability. We prove from a theoretical perspective that under reasonable conditions MixKD gives rise to a smaller gap between the error and the empirical error. Experiments under a limited-data setting and ablation studies further demonstrate the advantages of the proposed approach.
  arXiv  Detail & Related papers  (2020-11-01T18:47:51Z)

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.