FactUEP at SemEval-2026 Task 4: Structured Narrative Similarity Scoring with Aspect Decomposition and Weak-Signal Gating
TL;DR
An LLM-based sys-tem is introduced that operationalizes the three core dimen-sions—Abstract Theme, Course of Action, and Outcomes—via schema-constrained prompting to enforce structured outputs and alignment with the annotation protocol.
Abstract
This paper presents approach to narrative similarity prediction for SemEval-2026 Task 4 Track A. We introduce an LLM-based sys-tem that operationalizes the three core dimen-sions—Abstract Theme, Course of Action, and Outcomes—via schema-constrained prompting to enforce structured outputs and alignment with the annotation protocol. The sys-tem proceeds in three stages: structured aspect decomposition and scoring, weak-signal gating for low-confidence cases, and a targeted LLM-based tiebreak. The final model achieved near-human performance and ranked second on the Track A leaderboard.