@inproceedings{tripathi-etal-2026-ai,
    title = "{AI}-Monitors at {S}em{E}val-2026 Task 4: A Hybrid Embedding and {LLM} Ensemble for Narrative Similarity",
    author = "Tripathi, Vishnu  and
      -, Azad  and
      Joshi, Prakhar  and
      Sahoo, Pragyananda  and
      Kumar, Gaurav  and
      Arora, Piyush  and
      Mani, Neel",
    editor = "Kochmar, Ekaterina  and
      Ghosh, Debanjan  and
      North, Kai  and
      Komachi, Mamoru",
    booktitle = "Proceedings of the 20th {I}nternational {W}orkshop on {S}emantic {E}valuation (2026)",
    month = jul,
    year = "2026",
    address = "San Diego, California, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl-workshops/2026.semeval-1.271/",
    pages = "2139--2148",
    ISBN = "979-8-89176-414-9",
    abstract = "Narrative similarity requires reasoning over the deeper structural properties of stories - shared themes, causal progression, and outcomes - rather than surface-level lexical overlap. We describe AI-Monitors, our system for SemEval-2026 Task 4 (Track A), which determines which of two candidate stories is more narratively similar to a given anchor. We explore a progression of approaches - from embedding-based similarity to structured LLM prompting and ensemble construction - guided by four hypotheses about where narrative reasoning gains can be found. The final system achieves 75{\textbackslash}{\%} test accuracy on 400 instances, ranking 3rd out of 47 systems and approaching the individual human annotator ceiling of 78{\textbackslash}{\%}.Our key findings are: i) structured few-shot prompting substantially outperforms dense embedding similarity; ii) selecting ensemble components by how differently they make errors - rather than by accuracy alone - produces stronger predictions; and iii) how you describe an example to the model affects its predictions."
}