Decoupled prompt-guided mixture-of-experts dynamic distillation for multimodal recommendation

Lili WU; Renmin ZHANG; Bin ZHANG; Jincheng ZHANG; Xi CHEN; Yingjing QIAN

doi:10.1631/ENG.ITEE.2026.0160

Your Location：

Home >

Browse articles >

Decoupled prompt-guided mixture-of-experts dynamic distillation for multimodal recommendation

Regular Papers | Updated：2026-10-08

- Decoupled prompt-guided mixture-of-experts dynamic distillation for multimodal recommendation
- Decoupled prompt-guided mixture-of-experts dynamic distillation for multimodal recommendation
- ENGINEERING Information Technology & Electronic Engineering Vol. 27, Issue 9, (2026)
- Affiliations：
  
  School of Communication and Electronic Engineering, Jishou University, Jishou 416000, China
  Tencent Inc., Shenzhen 518057, China
  Ant Group Co., Ltd., Changsha 410000, China
  School of Electronic and Information Engineering, Huaihua University, Huaihua 418000, China
- Author bio：
  
  ✉Renmin ZHANG, rzhang1981@163.com
- Funds：
- DOI：10.1631/ENG.ITEE.2026.0160
  CLC： TP391.3
- Received：26 May 2026，
  
  Revised：2026-08-04，
  
  Published：01 September 2026
- Accepted：
Scan QR Code
Lili WU, Renmin ZHANG, Bin ZHANG, et al. Decoupled prompt-guided mixture-of-experts dynamic distillation for multimodal recommendation[J]. ENGINEERING Information Technology & Electronic Engineering, 2026, 27(9): 260160-15.
DOI：

Lili WU, Renmin ZHANG, Bin ZHANG, et al. Decoupled prompt-guided mixture-of-experts dynamic distillation for multimodal recommendation[J]. ENGINEERING Information Technology & Electronic Engineering, 2026, 27(9): 260160-15. DOI： 10.1631/ENG.ITEE.2026.0160.

摘要

Abstract

Multimodal recommendation aims to enrich preference modeling by leveraging visual and textual features. However

integrating high-dimensional pretrained features introduces substantial computational overhead. While knowledge distillation provides an effective compression strategy

existing frameworks face three intertwined challenges: rank bottlenecks caused by low-dimensional projections

cross-modal interference induced by shared fusion spaces

and optimization instability under static distillation temperatures. To address these issues

we propose ProMoE-DTS

a decoupled prompt-guided mixture-of-experts framework with dynamic temperature scheduling. Using an asymmetric teacher–student architecture

the teacher model leverages modality-aware soft prompts as semantic anchors to route heterogeneous features into parameter-disjoint expert networks

thereby alleviating cross-modal conflicts and resolving the rank bottlenecks. To ensure stable knowledge transfer

a feedback-driven dynamic temperature scheduler adaptively regulates the distillation intensity based on epoch-wise signals. This asymmetric design confines intensive multimodal operations to the offline teacher

leaving the online student model with a highly efficient

pure identifier-based structure. Extensive experiments on three benchmark datasets demonstrate that ProMoE-DTS improves Recall@20 by 2.24%–3.96% over state-of-the-art baselines

while requiring only 3.28%–3.55% of the teacher’s parameters.

关键词

Keywords

references

Bao KQ , Zhang JZ , Zhang Y , et al. , 2023 . TALLRec: an effective and efficient tuning framework to align large language model with recommendation . Proc 17 th ACM Conf on Recommender Systems , p. 1007 - 1014 . https://doi.org/10.1145/3604915.3608857 https://doi.org/10.1145/3604915.3608857

Chen JY , Zhang HW , He XN , et al. , 2017 . Attentive collaborative filtering: multimedia recommendation with item- and component-level attention . Proc 40 th Int ACM SIGIR Conf on Research and Development in Information Retrieval , p. 335 - 344 . https://doi.org/10.1145/3077136.3080797 https://doi.org/10.1145/3077136.3080797

Dai BQ , Du ZC , Zhu JM , et al. , 2024 . UniEmbedding: learning universal multi-modal multi-domain item embeddings via user-view contrastive learning . Proc 33 rd ACM Int Conf on Information and Knowledge Management , p. 4446 - 4453 . https://doi.org/10.1145/3627673.3680098 https://doi.org/10.1145/3627673.3680098

Deng JX , Wang SY , Cai K , et al. , 2025 . OneRec: unifying retrieve and rank with generative recommender and iterative preference alignment . https://arxiv.org/abs/2502.18965 https://arxiv.org/abs/2502.18965

Dong XM , Huang O , Thulasiraman P , et al. , 2023 . Improved knowledge distillation via teacher assistants for sentiment analysis . Proc IEEE Symp Series on Computational Intelligence , p. 300 - 305 . https://doi.org/10.1109/SSCI52147.2023.10371965 https://doi.org/10.1109/SSCI52147.2023.10371965

Du YP , Sun Z , Wang ZY , et al. , 2025 . Active large language model-based knowledge distillation for session-based recommendation . Proc 39 th AAAI Conf on Artificial Intelligence , p. 11607 - 11615 . https://doi.org/10.1609/aaai.v39i11.33263 https://doi.org/10.1609/aaai.v39i11.33263

Geng X , Zhang HW , Bian JW , et al. , 2015 . Learning image and user features for recommendation in social networks . Proc IEEE Int Conf on Computer Vision , p. 4274 - 4282 . https://doi.org/10.1109/ICCV.2015.486 https://doi.org/10.1109/ICCV.2015.486

He RN , McAuley J , 2016 . VBPR: visual Bayesian personalized ranking from implicit feedback . Proc 30 th AAAI Conf on Artificial Intelligence , p. 144 - 150 . https://doi.org/10.1609/aaai.v30i1.9973 https://doi.org/10.1609/aaai.v30i1.9973

He XN , Deng K , Wang X , et al. , 2020 . LightGCN: simplifying and powering graph convolution network for recommendation . Proc 43 rd Int ACM SIGIR Conf on Research and Development in Information Retrieval , p. 639 - 648 . https://doi.org/10.1145/3397271.3401063 https://doi.org/10.1145/3397271.3401063

Hinton G , Vinyals O , Dean J , 2015 . Distilling the knowledge in a neural network . https://arxiv.org/abs/1503.02531 https://arxiv.org/abs/1503.02531

Hu XH , Zhang HT , 2025 . Invariant representation learning in multimedia recommendation with modality alignment and model fusion . Entropy , 27 ( 1 ): 56 . https://doi.org/10.3390/e27010056 https://doi.org/10.3390/e27010056

Huo FS , Xu WC , Guo JC , et al. , 2024 . C 2 KD: bridging the modality gap for cross-modal knowledge distillation . Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition , p. 16006 - 16015 . https://doi.org/10.1109/CVPR52733.2024.01515 https://doi.org/10.1109/CVPR52733.2024.01515

Islam SU , Ahad JI , Rahman F , et al. , 2025 . Dynamic temperature scheduler for knowledge distillation . https://arxiv.org/abs/2511.13767 https://arxiv.org/abs/2511.13767

Jian M , Wang T , Yang MJ , et al. , 2026 . Hierarchy-aware multimodal distillation for recommendation . IEEE Trans Multim , 28 : 2279 - 2290 . https://doi.org/10.1109/TMM.2026.3651049 https://doi.org/10.1109/TMM.2026.3651049

Kang S , Kweon W , Lee D , et al. , 2023 . Distillation from heterogeneous models for top- K recommendation . Proc ACM Web Conf , p. 801 - 811 . https://doi.org/10.1145/3543507.3583209 https://doi.org/10.1145/3543507.3583209

Lan PX , Xu HY , Yang EN , et al. , 2025 . Efficient and effective prompt tuning via prompt decomposition and compressed outer product . Proc Conf of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies , p. 4406 - 4421 . https://doi.org/10.18653/v1/2025.naacl-long.225 https://doi.org/10.18653/v1/2025.naacl-long.225

Lee JW , Choi M , Lee J , et al. , 2019 . Collaborative distillation for top- N recommendation . Proc IEEE Int Conf on Data Mining , p. 369 - 378 . https://doi.org/10.1109/ICDM.2019.00047 https://doi.org/10.1109/ICDM.2019.00047

Li L , Zhang YF , Chen L , 2023 . Personalized prompt learning for explainable recommendation . ACM Trans Inform Syst , 41 ( 4 ): 103 . https://doi.org/10.1145/3580488 https://doi.org/10.1145/3580488

Liang JH , Zhao XY , Li MY , et al. , 2023 . MMMLP: multi-modal multilayer perceptron for sequential recommendations . Proc ACM Web Conf , p. 1109 - 1117 . https://doi.org/10.1145/3543507.3583378 https://doi.org/10.1145/3543507.3583378

Lin JH , Chen B , Wang HY , et al. , 2024 . ClickPrompt: CTR models are strong prompt generators for adapting language models to CTR prediction . Proc ACM Web Conf , p. 3319 - 3330 . https://doi.org/10.1145/3589334.3645396 https://doi.org/10.1145/3589334.3645396

Lin YF , He JY , 2026 . A reinforcement learning framework for multimodal recommendation explanation generation . Authorea . https://doi.org/10.22541/au.177368713.30806120/v1 https://doi.org/10.22541/au.177368713.30806120/v1

Liu F , Chen HL , Cheng ZY , et al. , 2023 . Disentangled multimodal representation learning for recommendation . IEEE Trans Multim , 25 : 7149 - 7159 . https://doi.org/10.1109/TMM.2022.3217449 https://doi.org/10.1109/TMM.2022.3217449

Liu MR , Zhang SX , Long C , 2025 . Facet-aware multi-head mixture-of-experts model for sequential recommendation . Proc 18 th ACM Int Conf on Web Search and Data Mining , p. 127 - 135 . https://doi.org/10.1145/3701551.3703552 https://doi.org/10.1145/3701551.3703552

Liu YF , Zhang KN , Ren XY , et al. , 2024 . AlignRec: aligning and training in multimodal recommendations . Proc 33 rd ACM Int Conf on Information and Knowledge Management , p. 1503 - 1512 . https://doi.org/10.1145/3627673.3679626 https://doi.org/10.1145/3627673.3679626

Loshchilov I , Hutter F , 2019 . Decoupled weight decay regularization . https://doi.org/10.48550/arXiv.1711.05101 https://doi.org/10.48550/arXiv.1711.05101

Ma WY , Xia HB , Liu Y , 2025 . DiffKD: collaborative graph diffusion with knowledge distillation for multimodal recommendation . J Intell Inform Syst , 63 ( 5 ): 1487 - 1510 . https://doi.org/10.1007/s10844-025-00946-4 https://doi.org/10.1007/s10844-025-00946-4

McAuley J , Targett C , Shi QF , et al. , 2015 . Image-based recommendations on styles and substitutes . Proc 38 th Int ACM SIGIR Conf on Research and Development in Information Retrieval , p. 43 - 52 . https://doi.org/10.1145/2766462.2767755 https://doi.org/10.1145/2766462.2767755

Mo F , Xiao L , Song QY , et al. , 2025 . FGCM: modality-behavior fusion model integrated with graph contrastive learning for multimodal recommendation . IEEE Multim , 32 ( 3 ): 27 - 37 . https://doi.org/10.1109/MMUL.2025.3542757 https://doi.org/10.1109/MMUL.2025.3542757

Nguyen NH , Nguyen TA , Nguyen T , et al. , 2024 . Towards efficient communication and secure federated recommendation system via low-rank training . Proc ACM Web Conf , p. 3940 - 3951 . https://doi.org/10.1145/3589334.3645702 https://doi.org/10.1145/3589334.3645702

Qiu RH , Wang S , Chen Z , et al. , 2021 . CausalRec: causal inference for visual debiasing in visually-aware recommendation . Proc 29 th ACM Int Conf on Multimedia , p. 3844 - 3852 . https://doi.org/10.1145/3474085.3475266 https://doi.org/10.1145/3474085.3475266

Reimers N , Gurevych I , 2019 . Sentence-BERT: sentence embeddings using Siamese BERT-networks . Proc Conf o n Empirical Methods in Natural Language Processing and the 9 th Int Joint Conf on Natural Language Processing , p. 3982 - 3992 . https://doi.org/10.18653/v1/D19-1410 https://doi.org/10.18653/v1/D19-1410

Rendle S , Freudenthaler C , Gantner Z , et al. , 2009 . BPR: Bayesian personalized ranking from implicit feedback . Proc 25 th Conf on Uncertainty in Artificial Intelligence , p. 452 - 461 .

Tang JX , Wang K , 2018 . Ranking distillation: learning compact ranking models with high performance for recommender system . Proc 24 th ACM SIGKDD Int Conf on Knowledge Discovery and Data Mining , p. 2289 - 2298 . https://doi.org/10.1145/3219819.3220021 https://doi.org/10.1145/3219819.3220021

Tian YJ , Zhang CX , Guo ZC , et al. , 2022 . NOSMOG: learning noise-robust and structure-aware MLPs on graphs . https://doi.org/10.48550/arXiv.2208.10010 https://doi.org/10.48550/arXiv.2208.10010

Wang J , Zhu L , Dai T , et al. , 2021 . Low-rank and sparse matrix factorization with prior relations for recommender systems . Appl Intell , 51 ( 6 ): 3435 - 3449 . https://doi.org/10.1007/s10489-020-02023-5 https://doi.org/10.1007/s10489-020-02023-5

Wang QY , Yin HZ , Wang H , et al. , 2019 . Enhancing collaborative filtering with generative augmentation . Proc 25 th ACM SIGKDD Int Conf on Knowledge Discovery and Data Mining , p. 548 - 556 . https://doi.org/10.1145/3292500.3330873 https://doi.org/10.1145/3292500.3330873

Wang X , He XN , Wang M , et al. , 2019 . Neural graph collaborative filtering . Proc 42 nd Int ACM SIGIR Conf on Research and Development in Information Retrieval , p. 165 - 174 . https://doi.org/10.1145/3331184.3331267 https://doi.org/10.1145/3331184.3331267

Wei RX , Lan JM , Li KK , et al. , 2025 . MFDB: multimodal feature fusion and dynamic behavior modeling for interactive recommendation systems . Knowl-Based Syst , 326 : 114047 . https://doi.org/10.1016/j.knosys.2025.114047 https://doi.org/10.1016/j.knosys.2025.114047

Wei W , Huang C , Xia LH , et al. , 2023 . Multi-modal self-supervised learning for recommendation . Proc ACM Web Conf , p. 790 - 800 . https://doi.org/10.1145/3543507.3583206 https://doi.org/10.1145/3543507.3583206

Wei W , Tang JB , Xia LH , et al. , 2024 . PromptMM: multi-modal knowledge distillation for recommendation with prompt-tuning . Proc ACM Web Conf , p. 3217 - 3228 . https://doi.org/10.1145/3589334.3645359 https://doi.org/10.1145/3589334.3645359

Wei YW , Wang X , Nie LQ , et al. , 2019 . MMGCN: multi-modal graph convolution network for personalized recommendation of micro-video . Proc 27 th ACM Int Conf on Multimedia , p. 1437 - 1445 . https://doi.org/10.1145/3343031.3351034 https://doi.org/10.1145/3343031.3351034

Wei YW , Wang X , Nie LQ , et al. , 2020 . Graph-refined convolutional network for multimedia recommendation with implicit feedback . Proc 28 th ACM Int Conf on Multimedia , p. 3541 - 3549 . https://doi.org/10.1145/3394171.3413556 https://doi.org/10.1145/3394171.3413556

Wei YW , Wang X , He XN , et al. , 2022 . Hierarchical user intent graph network for multimedia recommendation . IEEE Trans Multim , 24 : 2701 - 2712 . https://doi.org/10.1109/TMM.2021.3088307 https://doi.org/10.1109/TMM.2021.3088307

Xun JH , Zhang SY , Zhao Z , et al. , 2021 . Why do we click: visual impression-aware news recommendation . Proc 29 th ACM Int Conf on Multimedia , p. 3881 - 3890 . https://doi.org/10.1145/3474085.3475514 https://doi.org/10.1145/3474085.3475514

Yu HF , Qi ZY , Jang LK , et al. , 2024 . MMoE: enhancing multimodal models with mixtures of multimodal interaction experts . Proc Conf on Empirical Methods in Natural Language Processing , p. 10006 - 10030 . https://doi.org/10.18653/v1/2024.emnlp-main.558 https://doi.org/10.18653/v1/2024.emnlp-main.558

Zhang JH , Zhu YQ , Liu Q , et al. , 2021 . Mining latent structures for multimedia recommendation . Proc 29 th ACM Int Conf on Multimedia , p. 3872 - 3880 . https://doi.org/10.1145/3474085.3475259 https://doi.org/10.1145/3474085.3475259

Zhang ZJ , Liu SC , Yu JA , et al. , 2024 . M 3 oE: multi-domain multi-task mixture-of-experts recommendation framework . Proc 47 th Int ACM SIGIR Conf on Research and Development in Information Retrieval , p. 893 - 902 . https://doi.org/10.1145/3626772.3657686 https://doi.org/10.1145/3626772.3657686

Zhou X , Zhou HY , Liu Y , et al. , 2023 . Bootstrap latent representations for multi-modal recommendation . Proc ACM Web Conf , p. 845 - 854 . https://doi.org/10.1145/3543507.3583251 https://doi.org/10.1145/3543507.3583251

Zhu NJ , Ren YQ , Liu Y , et al. , 2025 . MMGCL: multi-scale and multi-channel graph contrastive learning for flight anomaly detection . Knowl-Based Syst , 329 : 114275 . https://doi.org/10.1016/j.knosys.2025.114275 https://doi.org/10.1016/j.knosys.2025.114275

Views

Downloads

CSCD

Alert me when the article has been cited

Submit

Tools

Publicity Resources

Adaptive practical fixed-time tracking control with prescribed performance for robotic manipulators

Relation-aware place recognition network for large-scale point clouds

Dynamic trend perception for fine-grained unmanned aerial vehicle anomaly detection in flight data

A comprehensive survey of data classification based on evidence theory

SigmaGait: silhouette-guided motion-augmentation module for depth-based gait recognition

Related Author

No data

Related Institution

No data

Chat

Address：Zhejiang University Press, 148 Tianmushan Road, Hangzhou, China Postal code：310028
Tel：+86-571-88273162 Email：fitee@zju.edu.cn
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰