@inproceedings{r-s-ulli-2026-team,
title = "Team {CV} at {S}em{E}val-2026 Task 4: Prompting {LLM}s and Benchmarking Embedding Models for Narrative Story Similarity",
author = "R S, Chandan Kumar and
Ulli, Vinay",
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://aclanthology.org/2026.semeval-1.374/",
pages = "2981--2985",
ISBN = "979-8-89176-414-9",
abstract = "This paper describes Team CV{'}s systems forSemEval-2026 Task 4: Narrative Story Sim-ilarity and Narrative Representation Learn-ing (Hatzel et al., 2026). For Track A (com-parative judgment), we explore five prompt-ing strategies{---}zero-shot, chain-of-thought,structured feature extraction, pairwise scor-ing, and few-shot{---}and QLoRA fine-tuningof smaller models. For Track B (narrativeembeddings), we benchmark twelve dedicatedtext embedding models of varying dimen-sionality (384{--}4096) spanning open-source(E5-Large-v2, BGE, GTE, Qwen3 Embed-ding) and closed-source (OpenAI, Gemini,Mistral) families, and fine-tune Qwen3 Em-bedding 4B on task-specific triples. Few-shot prompting with Qwen-2.5 7B (64.00{\%})outperforms all fine-tuned variants (best57.50{\%}) on Track A; scaling to LLaMA-3.3-70B yields 75.00{\%}. On Track B, Ope-nAI text-embedding-3-large (3072-d) achieves the best dev accuracy (67.00{\%}),while fine-tuning Qwen3 Embedding 4B(2560-d) on synthetic triples slightly de-creases accuracy. Our final submission{---}LLaMA-3.3-70B (3-shot) for Track A andtext-embedding-3-large for Track B{---}achieves 70.75{\%} and 64.50{\%}, exceeding theGPT-4o-mini and STORY-EMB baselines respec-tively."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="r-s-ulli-2026-team">
<titleInfo>
<title>Team CV at SemEval-2026 Task 4: Prompting LLMs and Benchmarking Embedding Models for Narrative Story Similarity</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chandan</namePart>
<namePart type="given">Kumar</namePart>
<namePart type="family">R S</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vinay</namePart>
<namePart type="family">Ulli</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 20th International Workshop on Semantic Evaluation (2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Kochmar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Debanjan</namePart>
<namePart type="family">Ghosh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kai</namePart>
<namePart type="family">North</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mamoru</namePart>
<namePart type="family">Komachi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-414-9</identifier>
</relatedItem>
<abstract>This paper describes Team CV’s systems forSemEval-2026 Task 4: Narrative Story Sim-ilarity and Narrative Representation Learn-ing (Hatzel et al., 2026). For Track A (com-parative judgment), we explore five prompt-ing strategies—zero-shot, chain-of-thought,structured feature extraction, pairwise scor-ing, and few-shot—and QLoRA fine-tuningof smaller models. For Track B (narrativeembeddings), we benchmark twelve dedicatedtext embedding models of varying dimen-sionality (384–4096) spanning open-source(E5-Large-v2, BGE, GTE, Qwen3 Embed-ding) and closed-source (OpenAI, Gemini,Mistral) families, and fine-tune Qwen3 Em-bedding 4B on task-specific triples. Few-shot prompting with Qwen-2.5 7B (64.00%)outperforms all fine-tuned variants (best57.50%) on Track A; scaling to LLaMA-3.3-70B yields 75.00%. On Track B, Ope-nAI text-embedding-3-large (3072-d) achieves the best dev accuracy (67.00%),while fine-tuning Qwen3 Embedding 4B(2560-d) on synthetic triples slightly de-creases accuracy. Our final submission—LLaMA-3.3-70B (3-shot) for Track A andtext-embedding-3-large for Track B—achieves 70.75% and 64.50%, exceeding theGPT-4o-mini and STORY-EMB baselines respec-tively.</abstract>
<identifier type="citekey">r-s-ulli-2026-team</identifier>
<location>
<url>https://aclanthology.org/2026.semeval-1.374/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>2981</start>
<end>2985</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Team CV at SemEval-2026 Task 4: Prompting LLMs and Benchmarking Embedding Models for Narrative Story Similarity
%A R S, Chandan Kumar
%A Ulli, Vinay
%Y Kochmar, Ekaterina
%Y Ghosh, Debanjan
%Y North, Kai
%Y Komachi, Mamoru
%S Proceedings of the 20th International Workshop on Semantic Evaluation (2026)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, USA
%@ 979-8-89176-414-9
%F r-s-ulli-2026-team
%X This paper describes Team CV’s systems forSemEval-2026 Task 4: Narrative Story Sim-ilarity and Narrative Representation Learn-ing (Hatzel et al., 2026). For Track A (com-parative judgment), we explore five prompt-ing strategies—zero-shot, chain-of-thought,structured feature extraction, pairwise scor-ing, and few-shot—and QLoRA fine-tuningof smaller models. For Track B (narrativeembeddings), we benchmark twelve dedicatedtext embedding models of varying dimen-sionality (384–4096) spanning open-source(E5-Large-v2, BGE, GTE, Qwen3 Embed-ding) and closed-source (OpenAI, Gemini,Mistral) families, and fine-tune Qwen3 Em-bedding 4B on task-specific triples. Few-shot prompting with Qwen-2.5 7B (64.00%)outperforms all fine-tuned variants (best57.50%) on Track A; scaling to LLaMA-3.3-70B yields 75.00%. On Track B, Ope-nAI text-embedding-3-large (3072-d) achieves the best dev accuracy (67.00%),while fine-tuning Qwen3 Embedding 4B(2560-d) on synthetic triples slightly de-creases accuracy. Our final submission—LLaMA-3.3-70B (3-shot) for Track A andtext-embedding-3-large for Track B—achieves 70.75% and 64.50%, exceeding theGPT-4o-mini and STORY-EMB baselines respec-tively.
%U https://aclanthology.org/2026.semeval-1.374/
%P 2981-2985
Markdown (Informal)
[Team CV at SemEval-2026 Task 4: Prompting LLMs and Benchmarking Embedding Models for Narrative Story Similarity](https://aclanthology.org/2026.semeval-1.374/) (R S & Ulli, SemEval 2026)
ACL