Generating humor is a complex and challenging task for Large Language Models (LLMs), requiring both linguistic creativity and strict adherence to constraints. This paper presents INF-rsrs, our solution for SemEval 2026 Task~1: Humor Generation, which tasks models with creating jokes from headlines and word pairs without labeled data. We propose a two-stage framework: a production stage and a selection stage. The production stage employs diverse model families and hyperparameter configurations to generate a wide range of candidate jokes, with each candidate generated by an LLM prompted in the role of a comedian under structured constraints to ensure relevance and humor. Our system was designed to substantiate our claim that the direct use of LLMs in creative works, such as humor generation, hits a hard ceiling that is inescapable through simple prompting. Our proposed system tied in first place in the task ranking, obtaining a top-tier performance.

Table Question Answering (TQA) is a challenging task that requires reasoning over structured data to extract accurate answers. This paper presents QleverAnswering-PUCRS, our submission to SemEval-2025 Task 8: DataBench, Question-Answering over Tabular Data. QleverAnswering-PUCRS is a modular multi-agent system that employs a structured approach to TQA. The approach revolves around breaking down the task into specialized agents, each dedicated to handling a specific aspect of the problem. Our system was evaluated on benchmark datasets and achieved competitive results, ranking mid-to-top positions in the SemEval-2025 competition. Despite these promising results, we identify areas for improvement, particularly in handling complex queries and nested data structures.

This paper describes the system submitted by our team (UFRGSent) to SemEval-2022 Task 10: Structured Sentiment Analysis. We propose a multilingual approach that relies on a Question Answering model to find tuples consisting of aspect, opinion, and holder. The approach starts from general questions and uses the extracted tuple elements to find the remaining components. Finally, we employ an aspect sentiment classification model to classify the polarity of the entire tuple. Despite our method being in a mid-rank position on SemEval competition, we show that the question-answering approach can achieve good coverage retrieving sentiment tuples, allowing room for improvements in the technique.

This paper describes the system submitted by our team (BabelEnconding) to SemEval-2020 Task 3: Predicting the Graded Effect of Context in Word Similarity. We propose an approach that relies on translation and multilingual language models in order to compute the contextual similarity between pairs of words. Our hypothesis is that evidence from additional languages can leverage the correlation with the human generated scores. BabelEnconding was applied to both subtasks and ranked among the top-3 in six out of eight task/language combinations and was the highest scoring system three times.