A multi-task approach to argument frame classification at variable granularity levels

TL;DR: An ensemble-based approach to multilingual framing detection achieved high performance, yielding a micro F 1 -score of 50.69% (rank 10) and a macro F 1 -score of 50.20% (rank 3) for English articles.

TL;DR: A novel unsupervised approach using large language models to generate indicative summaries for long discussions. The summaries serve as tables of contents and are generated using cluster labeling and frame classification techniques.

TL;DR: PAKT, a Perspectivized Argumentation Knowledge Graph and Tool, structures argumentative spaces across diverse topics, facilitating constructive discourse and informed decision making by analyzing structural patterns in deliberation processes and revealing multifaceted argumentation graphs.

TL;DR: A new global logbilinear regression model that combines the advantages of the two major model families in the literature: global matrix factorization and local context window methods and produces a vector space with meaningful substructure.

TL;DR: BERT as mentioned in this paper pre-trains deep bidirectional representations from unlabeled text by jointly conditioning on both left and right context in all layers, which can be fine-tuned with just one additional output layer to create state-of-the-art models for a wide range of tasks.

TL;DR: Reaching this goal would require a more self-con- scious determination by communication scholars to plumb other fields and feed back their studies to outside researchers, and enhance the theoretical rigor of communication scholarship proper.

TL;DR: In this paper, a gated recursive convolutional neural network (GRNN) was proposed to learn a grammatical structure of a sentence automatically, which performed well on short sentences without unknown words, but its performance degrades rapidly as the length of the sentence and the number of unknown words increase.