Molecular generation designs novel molecules with desired structures and properties. Global count constraints specify molecular size, elemental composition, and ring content without prescribing detailed structure. Property-conditioned generators steer generation toward such targets but do not guarantee exact satisfacti...
Xiang Wu, Rong-Hua Li, Quan-Qing Xu et al.· 0 citations
This work proposes TEAM, a modality-centric framework that constructs and encodes graph context with node-modality tokens rather than treating each node as a single interaction unit, and introduces the Modality-Source Context Transformer (MSCT), which encodes the selected node-modality tokens and coordinates interactio...
Yu-Shuo Li, Xun-Kai Li, Zheng-Yu Wu et al.· 0 citations
Geo-filtered approximate nearest-neighbor (GFANN) search retrieves the objects inside a query-specified geographic disk whose feature vectors are closest to a query vector, a core operation in location-aware Web search. A disk filter on a standard proximity graph invalidates out-of-disk routing vertices; on selective d...
Zhi-Qiu Zou, Rong-Hua Li, Qiang-Qiang Dai et al.· 0 citations
Graph Convolutional Networks (GCNs) are widely used for large graph-structured data, including social, citation, and e-commerce networks, but their deployment is constrained by irregular memory access and severe GPU workload imbalance. These challenges arise in two dimensions: width imbalance from power-law degree dist...
Lingling Zhang, Hang Zeng, Pengpeng Qiao et al.· 0 citations
Negative sampling is a key component of implicit collaborative filtering (CF), as it enables recommenders to effectively learn user preferences. Existing negative sampling methods mostly follow a two-stage paradigm: they first construct a candidate negative pool for each user and then select negative samples from the p...
Jia-Yi Wu, Zheng-Yu Wu, Xun-Kai Li et al.· 0 citations
GraphBind is a topology-driven approach that uses graph topology to bind rich modality information into a unified shared space that integrates structure and semantics, and adapts this space to discriminative and generative tasks through lightweight interfaces.
Xun-Kai Li, Chen-Xi Wan, Yin-Lin Zhu et al.· 0 citations
Resistance distance computation is a fundamental problem in graph data management. To avoid the high latency of online evaluation, recent work builds offline indexes. Existing approaches, however, fall into two families with complementary weaknesses: loop-erased random-walk methods (e.g., LEIndex) are efficient on fa...
Mei-Hao Liao, Bao Xing, Rong-Hua Li et al.· Proceedings of the ACM on Ma...· 0 citations
Density decomposition characterizes the multi-level dense structure of large networks and supports a wide range of graph mining applications. Given a graph
G = (V, E)
, it assigns each vertex an integral dense number (IDN) and produces a nested sequence of layers D
0
⊇ D
1
⊇ ... ⊇ D
p
that capture increasingl...
Ya-Long Zhang, Rong-Hua Li, Qi Zhang et al.· Proceedings of the ACM on Ma...· 0 citations
Multimodal graph learning faces a fundamental challenge: new classes may emerge after deployment, while models are trained with a fixed label space. Existing approaches typically detect unknown nodes and use LLMs to generate candidate class descriptions, but they do not determine whether existing classes are insufficie...
Ze-Kai Chen, Jia-Yang Xing, Xun Wu et al.· 0 citations
Diffusion planners exhibit strong capabilities in generating multimodal trajectories. However, existing methods primarily rely on expert demonstrations to fit trajectory distributions, learning statistical correlations among scenes, behaviors, and trajectories without explicitly modeling driving rules. In long-tail sce...
Jia-Xi Ye, Chun-Ji Lv, Guo-Ren Wang et al.· 0 citations
OmniVCBench is introduced, a figure-centric, source-traceable benchmark for the interpretation component of an AIVC, and AIVC-Judge is introduced, a task-conditioned MLLM-as-a-judge framework with category-specific, reference-aware rubrics for evaluating open-ended responses.
Man-Yu Li, Xun-Kai Li, Yong-Fu Xiong et al.· 0 citations
Graph-Optimized Multimodal Alignment (GOMA), which lets a jointly trained model support single-modality and dual-attribute retrieval through a task-specific readout, achieves state-of-the-art performance on all 14 primary measures against 14 external methods.