Researchers from UKP_Psycontrol have developed a system for SemEval-2026 Task 2, which focuses on predicting affective states and their changes from user-generated text. Their approach combined large language model prompting with a Maximum Entropy model and a neural regression model. While LLMs proved effective for current affect, the system found that recent affective trajectories were more predictive of short-term changes than textual content alone. The team achieved first place in both Subtask 1 and Subtask 2A of the competition. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Demonstrates LLM capabilities in affective computing and highlights the importance of temporal dynamics for predicting emotional shifts.
RANK_REASON This is a research paper detailing a system developed for a specific NLP task and its performance in a competition.
