Search results

Graphical Abstract Graphical Abstract For image description, please refer to the figure legend and surrounding text.

Figure 2 Taxonomic distribution of bacterial and fungal genomes by phylum. Bar charts show the number of genomes assigned to each phylum on a logarithmic scale. (a) Distribution of the 35 681 bacterial genomes, with Pseudomonadota, Bacillota, and Actinomycetota representing the most abundant phyla. (b) Distr

Graphical Abstract Graphical Abstract For image description, please refer to the figure legend and surrounding text.

Figure 8 UpSet plot analysing the trigger vocabulary composition across five data sources after the LLM enrichment step in dictionary construction. (1) The horizontal bars on the left represent the total vocabulary size of each dataset, comprising both translated vocabulary and generated aliases. (2) Similar

Figure 9 Effect of the similarity threshold on context-aware data augmentation. Bars indicate the number of generated training samples under each threshold after the first round of self-training, while lines show the F1 scores for trigger and argument extraction on the development set. All scores are reporte

Figure 3 Step-by-step workflow of the Search module. The interface allows users to construct complex, multi-layered queries. The numbered steps highlight the logical flow: (1) the user selects the main category. (2) A suitable sub-category is selected based on the intended screening strategy. (3) To deepen t

Figure 1 Qualitative and quantitative analysis of contextual scope in substance use NER. (a) Illustrative examples of heterogeneous contextual dependencies. (b) Aggregated token-level attention heatmap centred on the target mention (offset 0; grey band). The x-axis shows relative token offset (negative: prev

Figure 2 Overview of the I-NOWJ architecture. The framework consists of three main phases: adaptive context selection, knowledge-infused representation, and trigger–argument detection. For image description, please refer to the figure legend and surrounding text.

Figure 6 Ablation studies on context window sizes and context selection strategies. ACS denotes adaptive context selection. Panel (a) reports entity-level exact-match F1, and panel (b) reports token-level BIO F1. The bar charts show fixed context-window results, while the line plots show the corresponding F1

Figure 1 Overview of the ARTS-DB 2.0 web interface and search workflows. (a) The landing page provides two main entry points: the ‘Search’ module for exploring precomputed ARTS and FunARTS results using user-defined screening criteria, and the ‘Model-Oriented Search’ for gene-centred exploration across the d

Figure 4 Workflow and output of the Model-Oriented Search module. (a) (1) Designed for rapid, model-based exploration, the user first selects the target taxonomic domain (Bacteria or Fungus) to define the scope of the dataset. (2) A specific gene name (e.g.  gyrB) is then entered, and the search is executed.

Figure 1 Summary of the activities, deliverables, and growth of the AgBioData consortium during the NSF RCN grant from September 2021–June 2025. Graphic of AgBioData member composition, activities and deliverables.

Figure 3 An example of our adaptive context selection scheme. For image description, please refer to the figure legend and surrounding text.

Figure 4 Overview of the proposed context-aware data augmentation pipeline. Sentences containing labelled entities are selected and tokenized. ‘T’ and ‘A’ denote trigger and argument entities, respectively, while ‘W’ represents non-entity tokens. For image description, please refer to the figure legend an

Figure 5 Ablation analysis quantifying the performance degradation (in F1 score) when removing (w/o) or replacing (r/w) key components of the I-NOWJ framework. All values are reported in percentage points (%). For image description, please refer to the figure legend and surrounding text.

Figure 7 Venn diagrams illustrating the overlap of trigger vocabularies across resource categories. Numbers report base vocabularies after cross-lingual translation and filtering (before alias generation); overlaps are computed on these base sets. Solid circles indicate no alias expansion, whereas dashed out

Figure 2 Graphical User Interface of BiodiversityPMC. This screenshot displays the graphical user interface of BiodiversityPMC, allowing access to documents enriched with annotations visually integrated into the text. For image description, please refer to the figure legend and surrounding text.