Chao Song

JBHI Journal 2026 Journal Article

An Explainable Molecular Token Estimation Method for Knowledge-aware Drug-Drug Interaction Prediction

• Hui Yu
• Chao Song
• Jiahao Yuan
• Xinkun Li
• Xiao Zhang
• Yang Yang
• Zhe Yu
• Jian-Yu Shi

In molecular representation learning (MRL), tokens ( e. g. , atoms, motifs, and fingerprints) are the basic elements to represent molecules. It is a common practice by using various tokens to enhance the expressive power of Graph Neural Networks (GNNs) on molecular graphs. Although prior GNNs-based methods employing tokens achieve promising performances in drug-drug interaction (DDI) prediction, the influence of the token on the expressiveness of molecular embedding models remains underexplored. To bridge the gap, we provide an axiomatic definition of MRL from a frequency domain perspective, revealing that the model's performance is closely related to the number of tokens and deriving a theoretical upper bound of likelihood-based model convergency. Building on these insights, we propose SimMotifPro, a simple yet efficient motif-based method, for DDI prediction. Specifically, SimMotifPro uses a variant of DeeperGCN encoder and builds a motif-motif knowledge graph to capture motif interconnections. A Motif Ranker module is also introduced to decouple learned representations and differentiate the contributions of selected motifs. Empirically, we demonstrate that SimMotifPro adheres to the properties demonstrated in our theoretical upper bound and validate the general applicability of our theory across different methods. Furthermore, our approach achieves state-of-the-art performance on various benchmarks for DDI prediction. Our codes and checkpoints are available at https://github.com/siriusong/sim_motif_pro.

NeurIPS Conference 2025 Conference Paper

EnzyControl: Adding Functional and Substrate-Specific Control for Enzyme Backbone Generation

• Chao Song
• Zhiyuan Liu
• Han Huang
• Liang Wang
• Qiong Wang
• Jian-Yu Shi
• Hui Yu
• Yihang Zhou

Designing enzyme backbones with substrate-specific functionality is a critical challenge in computational protein engineering. Current generative models excel in protein design but face limitations in binding data, substrate-specific control, and flexibility for de novo enzyme backbone generation. To address this, we introduce EnzyBind, a dataset with 11, 100 experimentally validated enzyme-substrate pairs specifically curated from PDBbind. Building on this, we propose EnzyControl, a method that enables functional and substrate-specific control in enzyme backbone generation. Our approach generates enzyme backbones conditioned on MSA-annotated catalytic sites and their corresponding substrates, which are automatically extracted from curated enzyme-substrate data. At the core of EnzyControl is EnzyAdapter, a lightweight, modular component integrated into a pretrained motif-scaffolding model, allowing it to become substrate-aware. A two-stage training paradigm further refines the model's ability to generate accurate and functional enzyme structures. Experiments show that our EnzyControl achieves the best performance across structural and functional metrics on EnzyBind and EnzyBench benchmarks, with particularly notable improvements of 13% in designability and 13% in catalytic efficiency compared to the baseline models. The code is released at https: //github. com/Vecteur-libre/EnzyControl.

AAAI Conference 2025 Conference Paper

Integrating Personalized Spatio-Temporal Clustering for Next POI Recommendation

• Chao Song
• Zheng Ren
• Li Lu

Location-Based Social Networks (LBSNs) offer a rich dataset of user activity at Points-of-Interest (POIs), making next POI recommendation a key task. Traditional algorithms face challenges due to broad searching scopes, affecting recommendation accuracy. Users tend to visit nearby POIs and show temporal concentration in their activities, reflecting personalized spatio-temporal clustering. However, individual user data may be insufficient to capture these clustering effects for personalized recommendations. In this paper, we propose an integrated Personalized Spatio-Temporal Clustering Model (iPCM) for next POI recommendation. The model learns this kind of personalized spatio-temporal clustering effect by using global historical trajectory data in conjunction with user feature embeddings. It integrates the features of personalized spatio-temporal clustering with the user's trajectory, and completes the user's POI recommendation through a Transformer encoding and MLP decoding. To enhance the accuracy of predictions, we add a module of probability adjustment. The experimental results on multiple datasets show that with the help of personalized spatio-temporal clustering, the proposed iPCM is superior to existing methods in various evaluation metrics.

EAAI Journal 2024 Journal Article

Non-autoregressive transformer with fine-grained optimization for user-specified indoor layout

• Chao Song
• Jie Wang
• Shujie Chen
• Haidong Li
• Zhaoyi Jiang
• Bailin Yang

IJCAI Conference 2024 Conference Paper

Pre-training General User Representation with Multi-type APP Behaviors

• Yuren Zhang
• Min Hou
• Kai Zhang
• Yuqing Yuan
• Chao Song
• Zhihao Ye
• Enhong Chen
• Yang Yu

In numerous user-centric services on mobile applications (apps), accurately mining user interests and generating effective user representations are paramount. Traditional approaches, which often involve training task-specific user representations, are becoming increasingly impractical due to their high computational costs and limited adaptability. This paper introduces a novel solution to this challenge: the Multi-type App-usage Fusion Network (MAFN). MAFN innovatively pre-trains universal user representations, leveraging multi-type app behaviors to overcome key limitations in existing methods. We address two primary challenges: 1) the varying frequency of user behaviors (ranging from low-frequency actions like (un)installations to high-frequency yet insightful app launches); and 2) the integration of multi-type behaviors to form a cohesive representation. Our approach involves the creation of novel pre-training tasks that harness self-supervised signals from diverse app behaviors, capturing both long-term and short-term user interests. MAFN's unique fusion approach effectively amalgamates these interests into a unified vector space, facilitating the development of a versatile, general-purpose user representation. With a practical workflow, extensive experiments with three typical downstream tasks on real-world datasets verify the effectiveness of our approach.

NeurIPS Conference 2023 Conference Paper

AdaptSSR: Pre-training User Model with Augmentation-Adaptive Self-Supervised Ranking

• Yang Yu
• Qi Liu
• Kai Zhang
• Yuren Zhang
• Chao Song
• Min Hou
• Yuqing Yuan
• Zhihao Ye

User modeling, which aims to capture users' characteristics or interests, heavily relies on task-specific labeled data and suffers from the data sparsity issue. Several recent studies tackled this problem by pre-training the user model on massive user behavior sequences with a contrastive learning task. Generally, these methods assume different views of the same behavior sequence constructed via data augmentation are semantically consistent, i. e. , reflecting similar characteristics or interests of the user, and thus maximizing their agreement in the feature space. However, due to the diverse interests and heavy noise in user behaviors, existing augmentation methods tend to lose certain characteristics of the user or introduce noisy behaviors. Thus, forcing the user model to directly maximize the similarity between the augmented views may result in a negative transfer. To this end, we propose to replace the contrastive learning task with a new pretext task: Augmentation-Adaptive SelfSupervised Ranking (AdaptSSR), which alleviates the requirement of semantic consistency between the augmented views while pre-training a discriminative user model. Specifically, we adopt a multiple pairwise ranking loss which trains the user model to capture the similarity orders between the implicitly augmented view, the explicitly augmented view, and views from other users. We further employ an in-batch hard negative sampling strategy to facilitate model training. Moreover, considering the distinct impacts of data augmentation on different behavior sequences, we design an augmentation-adaptive fusion mechanism to automatically adjust the similarity order constraint applied to each sample based on the estimated similarity between the augmented views. Extensive experiments on both public and industrial datasets with six downstream tasks verify the effectiveness of AdaptSSR.

TIST Journal 2023 Journal Article

Enough Waiting for the Couriers: Learning to Estimate Package Pick-up Arrival Time from Couriers’ Spatial-Temporal Behaviors

• Haomin Wen
• Youfang Lin
• Fan Wu
• Huaiyu Wan
• Zhongxiang Sun
• Tianyue Cai
• Hongyu Liu
• Shengnan Guo

In intelligent logistics systems, predicting the Estimated Time of Pick-up Arrival (ETPA) of packages is a crucial task, which aims to predict the courier’s arrival time to all the unpicked-up packages at any time. Accurate prediction of ETPA can help systems alleviate customers’ waiting anxiety and improve their experience. We identify three main challenges of this problem. First, unlike the travel time estimation problem in other fields like ride-hailing, the ETPA task is distinctively a multi-destination and path-free prediction problem. Second, an intuitive idea for solving ETPA is to predict the pick-up route and then the time in two stages. However, it is difficult to accurately and efficiently predict couriers’ future routes in the route prediction step since their behaviors are affected by multiple complex factors. Third, furthermore, in the time prediction step, the requirement for providing a courier’s all unpicked-up packages’ ETPA at once in real time makes the problem even more challenging. To tackle the preceding challenges, we propose RankETPA, which integrates the route inference into the ETPA prediction. First, a learning-based pick-up route predictor is designed to learn the route-ranking strategies of couriers from their massive spatial-temporal behaviors. Then, a spatial-temporal attention-based arrival time predictor is designed for real-time ETPA inference via capturing the spatial-temporal correlations between the unpicked-up packages. Extensive experiments on two real-world datasets and a synthetic dataset demonstrate that RankETPA achieves significant performance improvement against the baseline models.

TIST Journal 2022 Journal Article

DeepRoute+: Modeling Couriers’ Spatial-temporal Behaviors and Decision Preferences for Package Pick-up Route Prediction

• Haomin Wen
• Youfang Lin
• Huaiyu Wan
• Shengnan Guo
• Fan Wu
• Lixia Wu
• Chao Song
• Yinghui Xu

Over 10 billion packages are picked up every day in China. A fundamental task raised in the emerging intelligent logistics systems is the couriers’ package pick-up route prediction, which is beneficial for package dispatching, arrival-time estimation and overdue-risk evaluation, by leveraging the predicted routes to improve those downstream tasks. In the package pick-up scene, the decision-making of a courier is affected by strict spatial-temporal constraints (e.g., package location, promised pick-up time, current time, and courier’s current location). Furthermore, couriers have different decision preferences on various factors (e.g., time factor, distance factor, and balance of both), based on their own perception of the environments and work experience. In this article, we propose a novel model, named DeepRoute+, to predict couriers’ future package pick-up routes according to the couriers’ decision experience and preference learned from the historical behaviors. Specifically, DeepRoute+ consists of three layers: (1) The representation layer produces experience- and preference-aware representations for the unpicked-up packages, in which a decision preference module can dynamically adjust the importance of factors that affects the courier’s decision under the current situation. (2) The transformer encoder layer encodes the representations of packages while considering the spatial-temporal correlations among them. (3) The attention-based decoder layer uses the attention mechanism to generate the whole pick-up route recurrently. Experiments on a real-world logistics dataset demonstrate the state-of-the-art performance of our model.

AAAI Conference 2020 Conference Paper

Enhancing Personalized Trip Recommendation with Attractive Routes

• Jiqing Gu
• Chao Song
• Wenjun Jiang
• Xiaomin Wang
• Ming Liu

Personalized trip recommendation tries to recommend a sequence of point of interests (POIs) for a user. Most of existing studies search POIs only according to the popularity of POIs themselves. In fact, the routes among the POIs also have attractions to visitors, and some of these routes have high popularity. We term this kind of route as Attractive Route (AR), which brings extra user experience. In this paper, we study the attractive routes to improve personalized trip recommendation. To deal with the challenges of discovery and evaluation of ARs, we propose a personalized Trip Recommender with POIs and Attractive Route (TRAR). It discovers the attractive routes based on the popularity and the Gini coefficient of POIs, then it utilizes a gravity model in a category space to estimate the rating scores and preferences of the attractive routes. Based on that, TRAR recommends a trip with ARs to maximize user experience and leverage the tradeoff between the time cost and the user experience. The experimental results show the superiority of TRAR compared with other state-of-the-art methods.

AAAI Conference 2020 Conference Paper

Spatial-Temporal Synchronous Graph Convolutional Networks: A New Framework for Spatial-Temporal Network Data Forecasting

• Chao Song
• Youfang Lin
• Shengnan Guo
• Huaiyu Wan

Spatial-temporal network data forecasting is of great importance in a huge amount of applications for traffic management and urban planning. However, the underlying complex spatial-temporal correlations and heterogeneities make this problem challenging. Existing methods usually use separate components to capture spatial and temporal correlations and ignore the heterogeneities in spatial-temporal data. In this paper, we propose a novel model, named Spatial-Temporal Synchronous Graph Convolutional Networks (STSGCN), for spatial-temporal network data forecasting. The model is able to effectively capture the complex localized spatial-temporal correlations through an elaborately designed spatial-temporal synchronous modeling mechanism. Meanwhile, multiple modules for different time periods are designed in the model to effectively capture the heterogeneities in localized spatialtemporal graphs. Extensive experiments are conducted on four real-world datasets, which demonstrates that our method achieves the state-of-the-art performance and consistently outperforms other baselines.

AAAI Conference 2019 Conference Paper

Attention Based Spatial-Temporal Graph Convolutional Networks for Traffic Flow Forecasting

• Shengnan Guo
• Youfang Lin
• Ning Feng
• Chao Song
• Huaiyu Wan

Forecasting the traffic flows is a critical issue for researchers and practitioners in the field of transportation. However, it is very challenging since the traffic flows usually show high nonlinearities and complex patterns. Most existing traffic flow prediction methods, lacking abilities of modeling the dynamic spatial-temporal correlations of traffic data, thus cannot yield satisfactory prediction results. In this paper, we propose a novel attention based spatial-temporal graph convolutional network (ASTGCN) model to solve traffic flow forecasting problem. ASTGCN mainly consists of three independent components to respectively model three temporal properties of traffic flows, i. e. , recent, daily-periodic and weekly-periodic dependencies. More specifically, each component contains two major parts: 1) the spatial-temporal attention mechanism to effectively capture the dynamic spatialtemporal correlations in traffic data; 2) the spatial-temporal convolution which simultaneously employs graph convolutions to capture the spatial patterns and common standard convolutions to describe the temporal features. The output of the three components are weighted fused to generate the final prediction results. Experiments on two real-world datasets from the Caltrans Performance Measurement System (PeMS) demonstrate that the proposed ASTGCN model outperforms the state-of-the-art baselines.