PromptMM: Multi-Modal Knowledge Distillation for Recommendation with Prompt-Tuning
Authors:
- Wei WeiWWWei Wei
- Jiabin TangJTJiabin Tang
- Yangqin JiangYJYangqin Jiang
- Lianghao XiaLXLianghao Xia
- Chao HuangCHChao Huang
Preprint
Abstract
Multimedia online platforms (e.g., Amazon, TikTok) have greatly benefited from the incorporation of multimedia (e.g., visual, textual, and acoustic) content into their personal recommender systems. These modalities provide intuitive semantics that facilitate modality-aware user preference modeling. However, two key challenges in multi-modal recommenders remain unresolved: i) The introduction of multi-modal encoders with a large number of additional parameters causes overfitting, given high-dimensional multi-modal features provided by extractors (e.g., ViT, BERT). ii) Side information inevitably introduces inaccuracies and redundancies, which skew the modality-interaction dependency from reflecting true user preference. To tackle these problems, we propose to simplify and empower recommenders through Multi-modal Knowledge Distillation (PromptMM) with the prompt-tuning that enables adaptive quality distillation. Specifically, PromptMM conducts model compression through distilling u-i edge relationship and multi-modal node content from cumbersome teachers to relieve students from the additional feature reduction parameters. To bridge the semantic gap between multi-modal context and collaborative signals for empowering the overfitting teacher, soft prompt-tuning is introduced to perform student task-adaptive. Additionally, to adjust the impact of inaccuracies in multimedia data, a disentangled multi-modal list-wise distillation is developed with modality-aware re-weighting mechanism. Experiments on real-world data demonstrate PromptMM's superiority over existing techniques. Ablation tests confirm the effectiveness of key components. Additional tests show the efficiency and effectiveness.Keywords
All Related Versions
This publication has 21 references indexed in Scilit:
- Knowledge Graph Contrastive Learning Based on Relation-Symmetrical StructureIEEE Transactions on Knowledge and Data Engineering, 2023
- Latent Structure Mining With Contrastive Modality Fusion for Multimedia RecommendationIEEE Transactions on Knowledge and Data Engineering, 2022
- Disentangled Multimodal Representation Learning for RecommendationIEEE Transactions on Multimedia, 2022
- On-Device Next-Item Recommendation with Self-Supervised Knowledge Distillation2022
- Selective Fairness in Recommendation via Prompts2022
- Self-Supervised Learning for Multimedia RecommendationIEEE Transactions on Multimedia, 2022
- DualGNN: Dual Graph Neural Network for Multimedia RecommendationIEEE Transactions on Multimedia, 2021
- Hierarchical User Intent Graph Network for Multimedia RecommendationIEEE Transactions on Multimedia, 2021
- On the optimization landscape of tensor decompositionsMathematical Programming, 2020
- Neural Graph Collaborative Filtering2019