Search

Supplemental material, sj-docx-1-taj-10.1177_20406223221112563 for Evidence of nonsurgical treatment for polycystic liver disease by Jeong-Ju Yoo, Hye In Jo, Eun-Ae Jung, Jae Seung Lee, Sang Gyune Kim, Young Seok Kim and Beom Kyung Kim in Therapeutic Advances in Chronic Disease

Supplemental material, sj-docx-1-taj-10.1177_20406223221112563 for Evidence of nonsurgical treatment for polycystic liver disease by Jeong-Ju Yoo, Hye In Jo, Eun-Ae Jung, Jae Seung Lee, Sang Gyune Kim, Young Seok Kim and Beom Kyung Kim in Therapeutic Advances in Chronic Disease

Supplemental material, sj-docx-1-tct-10.1177_15330338211041212 for Clinical Outcome of Salvage Radiotherapy for Locoregional Clinical Recurrence After Radical Prostatectomy by Sung Uk Lee, Kwan Ho Cho, Jin Ho Kim, Young Seok Kim, Taek-Keun Nam, Jae-Sung Kim, Jaeho Cho, Seo Hee Choi, Su Jung Shim, Jin Hee Kim and Ah Ram Chang in Technology in Cancer Research & Treatment

Supplemental material, sj-docx-1-tct-10.1177_15330338211041212 for Clinical Outcome of Salvage Radiotherapy for Locoregional Clinical Recurrence After Radical Prostatectomy by Sung Uk Lee, Kwan Ho Cho, Jin Ho Kim, Young Seok Kim, Taek-Keun Nam, Jae-Sung Kim, Jaeho Cho, Seo Hee Choi, Su Jung Shim, Jin Hee Kim and Ah Ram Chang in Technology in Cancer Research & Treatment

Graphs are the natural data structure to represent relational and structural information in many domains. To cover the broad range of graph-data applications including graph classification as well as graph generation, it is desirable to have a general and flexible model consisting of an encoder and a decoder that can handle graph data. Although the representative encoder-decoder model, Transformer, shows superior performance in various tasks especially of natural language processing, it is not immediately available for graphs due to their non-sequential characteristics. To tackle this incompatibility, we propose GRaph-Aware Transformer (GRAT), the first Transformer-based model which can encode and decode whole graphs in end-to-end fashion. GRAT is featured with a self-attention mechanism adaptive to the edge information and an auto-regressive decoding mechanism based on the two-path approach consisting of sub-graph encoding path and node-and-edge generation path for each decoding step. We empirically evaluated GRAT on multiple setups including encoder-based tasks such as molecule property predictions on QM9 datasets and encoder-decoder-based tasks such as molecule graph generation in the organic molecule synthesis domain. GRAT has shown very promising results including state-of-the-art performance on 4 regression tasks in QM9 benchmark.

The microelectronic industry has shown an increasing interest in TiN as a diffusion barrier for aluminum and copper metallization in ultra-large-scale integration (ULSI) devices because of its low electrical resistivity, high melting point, thermal stability, and good adhesion properties [1,2]. It is also used as a diffusion barrier for capacitor electrodes for high-k dielectric materials, such as Ta2O5 or BST [3], and is used as both a barrier against WF6 attack and a nucleation layer in the chemical vapor deposition (CVD) W plug filling process [4]. Both physical vapor deposition (PVD) and chemical vapor deposition (CVD) techniques have been applied to grow TiN film. Conventionally, TiN films have been deposited either by the reactive ion sputtering or by thermal nitridation of sputtered titanium. As integrated feature sizes have shrunk to submicron dimensions, the limitations of films produced by these physical techniques have become apparent. The poor step coverage of sputtered films, especially in submicron contact holes with