Now showing items 1-7 of 7

    • Another Dead End for Morphological Tags? Perturbed Inputs and Parsing 

      Muñoz-Ortiz, Alberto; Vilares, David (Association for Computational Linguistics, 2023-07)
      [Absctract]: The usefulness of part-of-speech tags for parsing has been heavily questioned due to the success of word-contextualized parsers. Yet, most studies are limited to coarse-grained tags and high quality written ...
    • Assessment of Pre-Trained Models Across Languages and Grammars 

      Muñoz-Ortiz, Alberto; Vilares, David; Gómez-Rodríguez, Carlos (Association for Computational Linguistics, 2023-11)
      [Absctract]: We present an approach for assessing how multilingual large language models (LLMs) learn syntax in terms of multi-formalism syntactic structures. We aim to recover constituent and dependency structures by ...
    • Contrasting Linguistic Patterns in Human and LLM-Generated News Text 

      Muñoz-Ortiz, Alberto; Gómez-Rodríguez, Carlos; Vilares, David (Springer, 2024)
      [Abstract]: We conduct a quantitative analysis contrasting human-written English news text with comparable large language model (LLM) output from six different LLMs that cover three different families and four sizes in ...
    • Cross-lingual Inflection as a Data Augmentation Method for Parsing 

      Muñoz-Ortiz, Alberto; Gómez-Rodríguez, Carlos; Vilares, David (Association for Computational Linguistics, 2022-05)
      [Absctract]: We propose a morphology-based method for low-resource (LR) dependency parsing. We train a morphological inflector for target LR languages, and apply it to related rich-resource (RR) treebanks to create ...
    • Demographic Background Prompting Does Not Affect Linguistic Features on LLM-Generated News Texts 

      Gómez-Rodríguez, Carlos; Vilares, David; Muñoz-Ortiz, Alberto (2024)
      "We explored if implicit demographic information in prompts for large language models (LLMs) influences the linguistic features of generated text. Two LLMs were prompted to write news articles based on a title and summary, ...
    • Not All Linearizations Are Equally Data-Hungry in Sequence Labeling Parsing 

      Muñoz-Ortiz, Alberto; Strzyz, Michalina; Vilares, David (INCOMA Ltd., 2021-09)
      [Absctract]: Different linearizations have been proposed to cast dependency parsing as sequence labeling and solve the task as: (i) a head selection problem, (ii) finding a representation of the token arcs as bracket ...
    • Parsing linearizations appreciate PoS tags - but some are fussy about errors 

      Muñoz-Ortiz, Alberto; Anderson, Mark; Vilares, David; Gómez-Rodríguez, Carlos (Association for Computational Linguistics, 2022-11)
      [Absctract]: PoS tags, once taken for granted as a useful resource for syntactic parsing, have become more situational with the popularization of deep learning. Recent work on the impact of PoS tags on graph- and ...