Temporal link prediction with temporal graph neural networks (TGNNs) is increasingly used to model spatio-temporal dependencies in temporal graphs and to forecast future interactions among entities. Existing sampling-based training methods typically rely on random negative sampling and pointwise loss formulations, which often lead to suboptimal convergence and limited generalization due to low-quality negative samples. We propose ATNSF, a temporal graph learning framework with a hybrid negative sampling strategy that uses a portion of historical edges as hard negatives. For efficiency, we design an asynchronous parallel training pipeline for scalable optimization and introduce a pairwise sampled softmax loss that contrasts each positive instance with a batch of negatives to learn more discriminative representations. Finally, we theoretically show that jointly designing the loss function and negative sampling strategy is crucial for improving performance and generalization. Extensive experiments across six temporal graph datasets demonstrate that ATNSF improves the average AP from 0.724 to 0.827 (+0.103). Remarkably, it also accelerates training by 1.37× to 6.85×, achieving a 2.57× geometric mean speedup. The source code of this paper can be found at https://github.com/yongqiu-star/ATNSF.
Yong-Chun Jiang, Heng Zhang, Jian Gao et al.· Proceedings of the Thirty-Fi...· 0 citations
De novo protein design is pivotal for revolutionizing protein engineering and advancing life sciences. Protein co-design aims to simultaneously create a novel protein sequence and structure with tailored functions, addressing the insufficient consistency between sequence and structure of two-stage design. Current AI-assisted protein co-design approaches primarily rely on protein sequence and structure information. However, they face two major challenges in limited integration of diverse biological knowledge and insufficient understanding of 'sequence-structure-function' relations, hindering the discovery of functional and diverse proteins in de novo design. To address these challenges, we propose a Protein sequence–structure–function Consistency Design model empowered by Natural Language function description, dubbed ProtcdNl, which expands the protein design space and ensures alignment with function-aware framework. Concretely, ProtcdNl contains two core components: i) the triple-coupled collaborative encoder, which achieves implicit alignment via joint latent space constraints, precisely maps natural language functional semantics to geometric protein motifs, and ii) a function-aware equivariant decoder, which endows the model with functional awareness while strictly maintaining the symmetry of molecular dynamics. Extensive experiments on our proposed ProtSSGT corpus demonstrate that ProtcdNl effectively mines the latent associations between functional semantics and protein geometry, achieving the design of novel, diverse proteins with high functional fidelity.
Ming Yang, Xin Zheng, Yi Li et al.· Proceedings of the 32nd ACM...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
