5 accepted papers
Data perspectivism goes beyond majority vote label aggregation by recognizing various perspectives as legitimate ground truths.However, current evaluation practices remain fragmented, making it difficult to compare perspectivist approaches and analyze their impact on different users and demographic
Large language models readily memorize arbitrary training instances, such as label noise, yet they perform strikingly well on reasoning tasks. In this work, we investigate how language models memorize label noise, and why such memorization in many cases does not heavily affect generalizable reasonin
Generating ironic content is challenging: it requires a nuanced understanding of context and implicit references and balancing seriousness and playfulness. Moreover, irony is highly subjective and can depend on various factors, such as social, cultural, or generational aspects. This paper explores w…
Recently, several scholars have contributed to the growth of a new theoretical framework in NLP called perspectivism. This approach aimsto leverage data annotated by different individuals to model diverse perspectives that affect their opinions on subjective phenomena such as irony. In this context,…
Research in the field of NLP has recently focused on the variability that people show in selecting labels when performing an annotation task. Exploiting disagreements in annotations has been shown to offer advantages for accurate modelling and fair evaluation. In this paper, we propose a strongly pe…