Recent years have witnessed the flourishing of self-supervised learning (SSL) on heterogeneous graphs, e.g., masked autoencoder (MAE)-based methods. However, existing MAE-based methods mostly suffer from three issues, including the prohibitive computational cost of constructing meta-path-induced subgraphs, the limitati...
Hao Geng, De-Qing Wang, Li-Hao Ge et al.· IEEE Transactions on Neural...· 0 citations
LoG-Reranker is proposed, a local-to-global sentence-level graph reranking framework for scientific synthesis that consistently outperforms competitive rerankers, yielding more reliable rankings and improving the quality of generated synthesis.
AdaVSkip is proposed, which equips each layer with two lightweight routers that independently determine whether visual tokens pass through by or skip the self-attention and MLP modules, and maintains strong task performance with substantially less computation.
Yu-Yao Sun, Tao Deng, Shuang-Hua Li et al.· arXiv.org· 0 citations
Middle-layer Attention Prediction (MAP), which uses Question Contrastive Teacher Selection to identify a sample-specific teacher layer by contrasting attention under the original and reference questions, and distills attention from the selected layer into a lightweight predictor that estimates visual token importance f...
Yu-Yao Sun, Tao Deng, Shuang-Hua Li et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
