arXiv subjects

Agnese Daffara

Publications and source records attributed to Agnese Daffara.

4 recordsLinked to original sources

arXiv2026

Structuring the Space of Perspectives

The same event can be reported from different perspectives depending on the experiences, background, and beliefs of the writer or speaker. A variety of NLP areas engage with perspectives, spanning from text analysis to algorithm optimization. A wide range of operative concepts (such as stances, sentiment, frames, and arguments) has been used to capture perspectives in texts, however the precise relationships among those concepts remain unclear. Arguably, a deeper theoretical understanding of these concepts would empower more effective research on perspectives. In this paper, we address this gap by reviewing the space of perspectives in NLP and defining a set of properties that help distinguishing perspective-related concepts. Our analysis leads us to posit a hierarchy which organizes these concepts linearly along a single axis. Finally, we show how this principled conceptual hierarchy can help researchers navigate the field and select operationalizations of perspective that align with their specific research objectives.

cs.CL↗

arXiv2025

Leveraging Media Frames to Improve Normative Diversity in News Recommendations

Click-based news recommender systems suggest users content that aligns with their existing history, limiting the diversity of articles they encounter. Recent advances in aspect-based diversification -- adding features such as sentiments or news categories (e.g. world, politics) -- have made progress toward diversifying recommendations in terms of perspectives. However, these approaches often overlook the role of news framing, which shapes how stories are told by emphasizing specific angles or interpretations. In this paper, we treat media frames as a controllable aspect within the recommendation pipeline. By selecting articles based on a diversity of frames, our approach emphasizes varied narrative angles and broadens the interpretive space recommended to users. In addition to introducing frame-based diversification method, our work is the first to assess the impact of a news recommender system that integrates frame diversity using normative diversity metrics: representation, calibration, and activation. Our experiments based on media frame diversification show an improvement in exposure to previously unclicked frames up to 50%. This is important because repeated exposure to the same frames can reinforce existing biases or narrow interpretations, whereas introducing novel frames broadens users' understanding of issues and perspectives. The method also enhances diversification across categorical and sentiment levels, thereby demonstrating that framing acts as a strong control lever for enhancing normative diversity.

cs.IR↗

arXiv2025

Generalizability of Media Frames: Corpus creation and analysis across countries

Frames capture aspects of an issue that are emphasized in a debate by interlocutors and can help us understand how political language conveys different perspectives and ultimately shapes people's opinions. The Media Frame Corpus (MFC) is the most commonly used framework with categories and detailed guidelines for operationalizing frames. It is, however, focused on a few salient U.S. news issues, making it unclear how well these frames can capture news issues in other cultural contexts. To explore this, we introduce FrameNews-PT, a dataset of Brazilian Portuguese news articles covering political and economic news and annotate it within the MFC framework. Through several annotation rounds, we evaluate the extent to which MFC frames generalize to the Brazilian debate issues. We further evaluate how fine-tuned and zero-shot models perform on out-of-domain data. Results show that the 15 MFC frames remain broadly applicable with minor revisions of the guidelines. However, some MFC frames are rarely used, and novel news issues are analyzed using general 'fall-back' frames. We conclude that cross-cultural frame use requires careful consideration.

cs.CL↗

arXiv2025

Fine-grained Fallacy Detection with Human Label Variation

We introduce Faina, the first dataset for fallacy detection that embraces multiple plausible answers and natural disagreement. Faina includes over 11K span-level annotations with overlaps across 20 fallacy types on social media posts in Italian about migration, climate change, and public health given by two expert annotators. Through an extensive annotation study that allowed discussion over multiple rounds, we minimize annotation errors whilst keeping signals of human label variation. Moreover, we devise a framework that goes beyond "single ground truth" evaluation and simultaneously accounts for multiple (equally reliable) test sets and the peculiarities of the task, i.e., partial span matches, overlaps, and the varying severity of labeling errors. Our experiments across four fallacy detection setups show that multi-task and multi-label transformer-based approaches are strong baselines across all settings. We release our data, code, and annotation guidelines to foster research on fallacy detection and human label variation more broadly.

cs.CL↗