The anonymization of clinical free-text records is a prerequisite for enabling the secondary use of healthcare data while preserving patient privacy. This challenge is particularly acute for Spanish clinical text, where annotated resources are scarce and practitioners lack clear empirical guidance on which technological approaches are more adequate to their particular restrictions and capabilities. In this work, we present a controlled comparative study of representative anonymization paradigms for Spanish clinical narratives, including a baseline rule-based approach, a general-purpose large language model under prompt-based inference, an off-the-shelf industrial NLP toolkit (spaCy) and comparable neural sequence labeling architectures. To ensure a fair and contamination-aware evaluation, particularly given the opacity of pretrained model training data, we introduce a synthetic clinical dataset. Recurrent neural network architectures, particularly the off-the-shelf spaCy toolkit, consistently achieve the best balance between effectiveness, computational efficiency, and deployment feasibility. We further observe that training task-specific embeddings end-to-end yields stronger generalization than incorporating pretrained representations. Although limited to Spanish and to representative instances of each paradigm, the study identifies stable performance tendencies across datasets. These results provide actionable guidance for institutions seeking to implement anonymization pipelines. This work contributes reproducible evaluation procedures and empirical evidence for privacy-preserving clinical NLP in Spanish.

Many approaches to Natural Language Processing tasks often treat them as single-step problems, where an agent receives an instruction, executes it, and is evaluated based on the final outcome. However, language is inherently interactive, as evidenced by the back-and-forth nature of human conversations. In light of this, we posit that human-AI collaboration should also be interactive, with humans monitoring the work of AI agents and providing feedback that the agent can understand and utilize. Further, the AI agent should be able to detect when it needs additional information and proactively ask for help. Enabling this scenario would lead to more natural, efficient, and engaging human-AI collaboration.In this paper, we investigate these directions using the challenging task established by the IGLU competition, an interactive grounded language understanding task in a MineCraft-like world. We delve into multiple types of help players can give to the AI to guide it and analyze the impact of this help on behavior, resulting in performance improvements and an end-to-end interactive system.

Despite the recent progress in language generation models, their outputs may not always meet user expectations. In this work, we study whether informational feedback in natural language can be leveraged to improve generation quality and user preference alignment. To this end, we consider factual consistency in summarization, the quality that the summary should only contain information supported by the input documents, as the user-expected preference. We collect a high-quality dataset, DeFacto, containing human demonstrations and informational natural language feedback consisting of corrective instructions, edited summaries, and explanations with respect to the factual consistency of the summary. Using our dataset, we study three natural language generation tasks: (1) editing a summary by following the human feedback, (2) generating human feedback for editing the original summary, and (3) revising the initial summary to correct factual errors by generating both the human feedback and edited summary. We show that DeFacto can provide factually consistent human-edited summaries and further insights into summarization factual consistency thanks to its informational natural language feedback. We further demonstrate that fine-tuned language models can leverage our dataset to improve the summary factual consistency, while large language models lack the zero-shot learning ability in our proposed tasks that require controllable text generation.

In this paper, we present a Wikipedia-based approach to develop resources for the legal domain. We establish a mapping between a legal domain ontology, LKIF (Hoekstra et al. 2007), and a Wikipedia-based ontology, YAGO (Suchanek et al. 2007), and through that we populate LKIF. Moreover, we use the mentions of those entities in Wikipedia text to train a specific Named Entity Recognizer and Classifier. We find that this classifier works well in the Wikipedia, but, as could be expected, performance decreases in a corpus of judgments of the European Court of Human Rights. However, this tool will be used as a preprocess for human annotation. We resort to a technique called “curriculum learning” aimed to overcome problems of overfitting by learning increasingly more complex concepts. However, we find that in this particular setting, the method works best by learning from most specific to most general concepts, not the other way round.