arXiv:2512.07068v1 Announce Type: new 
Abstract: Uniform Meaning Representation (UMR) is a novel graph-based semantic representation which captures the core meaning of a text, with flexibility incorporated into the annotation schema such that the breadth of the world's languages can be annotated (including low-resource languages). While UMR shows promise in enabling language documentation, improving low-resource language technologies, and adding interpretability, the downstream applications of UMR can only be fully explored when text-to-UMR parsers enable the automatic large-scale production of accurate UMR graphs at test time. Prior work on text-to-UMR parsing is limited to date. In this paper, we introduce two methods for English text-to-UMR parsing, one of which fine-tunes existing parsers for Abstract Meaning Representation and the other, which leverages a converter from Universal Dependencies, using prior work as a baseline. Our best-performing model, which we call SETUP, achieves an AnCast score of 84 and a SMATCH++ score of 91, indicating substantial gains towards automatic UMR parsing.

تم تقديم محلل جديد يسمى SETUP لتحويل الجمل الإنجليزية إلى تمثيل المعنى الموحد (UMR)، وهو تمثيل دلالي قائم على الرسوم البيانية يلتقط المعنى الأساسي للنصوص. يهدف هذا التطور إلى تحسين الإنتاج التلقائي للرسوم البيانية UMR، وهو أمر حاسم لمجموعة متنوعة من التطبيقات في معالجة اللغة. تبرز الدراسة طريقتين لتحقيق هذا التحليل، مستندة إلى أطر موجودة في تمثيل المعنى المجرد والاعتماد العالمي.

Se ha introducido un nuevo parser llamado SETUP para convertir oraciones en inglés a Representación de Significado Uniforme (UMR), una representación semántica basada en grafos que captura el significado central de los textos. Este desarrollo tiene como objetivo mejorar la generación automática de grafos UMR, lo cual es crucial para diversas aplicaciones en el procesamiento del lenguaje. La investigación destaca dos métodos para lograr este parsing, basándose en marcos existentes en Representación de Significado Abstracto y Dependencias Universales.

Un nouveau parseur nommé SETUP a été introduit pour convertir des phrases en anglais en Représentation de Sens Uniforme (UMR), une représentation sémantique basée sur des graphes qui capture le sens fondamental des textes. Ce développement vise à améliorer la génération automatique de graphes UMR, ce qui est crucial pour diverses applications dans le traitement des langues. La recherche met en évidence deux méthodes pour réaliser ce parsing, s'appuyant sur des cadres existants en Représentation de Sens Abstrait et Dépendances Universelles.

A new parser named SETUP has been introduced for converting English sentences into Uniform Meaning Representation (UMR), a graph-based semantic representation that captures the core meaning of texts. This development aims to enhance the automatic generation of UMR graphs, which is crucial for various applications in language processing. The research highlights two methods for achieving this parsing, building on existing frameworks in Abstract Meaning Representation and Universal Dependencies.

SETUP: Sentence-level English-To-Uniform Meaning Representation Parser

arXiv:2502.11973v3 Announce Type: replace 
Abstract: Uniform Meaning Representation (UMR) is a recently developed graph-based semantic representation, which expands on Abstract Meaning Representation (AMR) in a number of ways, in particular through the inclusion of document-level information and multilingual flexibility. In order to effectively adopt and leverage UMR for downstream tasks, efforts must be placed toward developing a UMR technological ecosystem. Though only a small amount of UMR annotations have been produced to date, in this work, we investigate the first approaches to producing text from multilingual UMR graphs. Exploiting the structural similarity between UMR and AMR graphs and the wide availability of AMR technologies, we introduce (1) a baseline approach which passes UMR graphs to AMR-to-text generation models, (2) a pipeline conversion of UMR to AMR, then using AMR-to-text generation models, and (3) a fine-tuning approach for both foundation models and AMR-to-text generation models with UMR data. Our best performing models achieve multilingual BERTscores of 0.825 for English and 0.882 for Chinese, a promising indication of the effectiveness of fine-tuning approaches for UMR-to-text generation even with limited UMR data.

أدت التطورات الأخيرة في تمثيل المعنى الموحد (UMR) إلى استكشاف طرق لتوليد النصوص من الرسوم البيانية UMR متعددة اللغات، مما يعزز قدرات التمثيل الدلالي في معالجة اللغة الطبيعية. تهدف هذه الأبحاث إلى تطوير نظام تكنولوجي حول UMR، بالاعتماد على الأطر الموجودة لتمثيل المعنى المجرد (AMR).

Los recientes avances en la Representación de Significado Uniforme (UMR) han llevado a la exploración de métodos para generar texto a partir de gráficos UMR multilingües, mejorando así las capacidades de representación semántica en el procesamiento del lenguaje natural. Esta investigación tiene como objetivo desarrollar un ecosistema tecnológico en torno a la UMR, basándose en los marcos existentes de la Representación de Significado Abstracto (AMR).

Des avancées récentes dans la représentation sémantique uniforme (UMR) ont conduit à l'exploration de méthodes pour générer du texte à partir de graphes UMR multilingues, améliorant ainsi les capacités de représentation sémantique dans le traitement du langage naturel. Cette recherche vise à développer un écosystème technologique autour de l'UMR, en s'appuyant sur les cadres existants de la représentation sémantique abstraite (AMR).

Recent advancements in Uniform Meaning Representation (UMR) have led to the exploration of methods for generating text from multilingual UMR graphs, enhancing the capabilities of semantic representation in natural language processing. This research aims to develop a technological ecosystem around UMR, building on the existing frameworks of Abstract Meaning Representation (AMR).

SETUP: Sentence-level English-To-Uniform Meaning Representation Parser

Was this article worth reading? Share it

Airparser

LucidQuery AI

Humanize AI

Supametas.AI

Uwear

Postugc

Ready to build your own newsroom?