Asa

This is the official implementation for our paper Recommendation of Mix-and-Match Clothing by Modeling Indirect Personal Compatibility, accepted by ICMR'23.

No description.

No description.

Research code for TacoRec, an AI model that studies multiple types of user behavior on e-commerce platforms at once to make more accurate product recommendations than single-behavior approaches.

This repo uses a combination of logits and feature distillation method to teach the PSPNet model of ResNet18 backbone with the PSPNet model of ResNet50 backbone. All the models are trained and tested on the PASCAL-VOC2012 dataset.

No description.

VBPR is a modified version with two simple BPR branches for compatibility modeling between a given top and a bottom to be matched. These branches are dedicated to measuring similarity in ID embeddings and visual latent features, employing the cosine function for all similarity metrics.

No description.

[WSDM'2024 Oral] "SSLRec: A Self-Supervised Learning Framework for Recommendation"

Modeling Multi-Relational Connectivity for Personalized Fashion Matching, ACM MM 2023

[KDD'22] Multi-Behavior Hypergraph-Enhanced Transformer for Next-Item Recommendation

(ICCV'21) Official code of "Dressing in Order: Recurrent Person Image Generation for Pose Transfer, Virtual Try-on and Outfit Editing."