Multiperspectivity as a Resource for Narrative Similarity Prediction

TL;DR

This paper explores incorporating multiple interpretive perspectives into narrative similarity prediction using an ensemble of diverse language model personas, improving accuracy and highlighting the importance of interpretive plurality in evaluation.

Contribution

It introduces a novel ensemble approach with diverse LLM personas to account for interpretive variability in narrative similarity tasks.

Findings

Ensemble accuracy reaches 0.705 on SemEval-2026 dataset.

Accuracy increases with ensemble size, following Condorcet Jury Theorem dynamics.

Gender-focused interpretive vocabulary negatively correlates with accuracy.

Abstract

Predicting narrative similarity can be understood as an inherently interpretive task: different, equally valid readings of the same text can produce divergent interpretations and thus different similarity judgments, posing a fundamental challenge for semantic evaluation benchmarks that encode a single ground truth. Rather than treating this multiperspectivity as a challenge to overcome, we propose to incorporate it in the decision making process of predictive systems. To explore this strategy, we created an ensemble of 31 LLM personas. These range from practitioners following interpretive frameworks to more intuitive, lay-style characters. Our experiments were conducted on the SemEval-2026 Task 4 dataset, where the system achieved an accuracy score of 0.705. Accuracy improves with ensemble size, consistent with Condorcet Jury Theorem-like dynamics under weakened independence. Practitioner personas perform worse individually but produce less correlated errors, yielding larger ensemble gains under majority voting. Our error analysis reveals a consistent negative association between gender-focused interpretive vocabulary and accuracy across all persona categories, suggesting either attention to dimensions not relevant for the benchmark or valid interpretations absent from the ground truth. This finding underscores the need for evaluation frameworks that account for interpretive plurality.