@inproceedings{thorat-etal-2026-plottwist,
title = "{P}lot{T}wist: A Creative Plot Generation Framework with Small Language Models",
author = "Thorat, Abhinav and
Kolla, Ravi Kumar and
Goel, Jyotin and
Kataria, Madhav and
Pedanekar, Niranjan",
editor = "Sousa-Silva, Rui and
Lopes Cardoso, Henrique and
Koponen, Maarit and
Pareja-Lora, Antonio",
booktitle = "Proceedings of the 3rd Workshop on Language Understanding in the Human-Machine Era ({LUHME} 2026)",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.luhme-1.13/",
pages = "144--172",
ISBN = "979-8-89176-471-2",
abstract = "Creative plot generation presents a fundamental challenge for language models: transforming a concise premise into a coherent narrative while satisfying multiple narrative constraints, including global coherence, character development, pacing, tone consistency, and emotional progression. Although recent Large Language Models (LLMs) demonstrate strong fluency on general-purpose tasks, they require preference alignment to perform well on domain-specific tasks such as creative plot generation. However, conducting such alignment at the scale of frontier LLMs is computationally prohibitive, significantly limiting accessibility and practical deployment. To address this, we present \textsc{PlotTwist}, a structured framework that enables Small Language Models (SLMs) with $\leq 3$B active parameters to generate high-quality, premise-conditioned plots competitive with frontier systems of vastly greater parameter scale. Our approach decomposes generation into three specialized components: (1) an Aspect Rating Reward Model, trained via a novel Positive{--}Negative prompting strategy; (2) a Mixture-of-Experts (MoE) plot generator aligned via Direct Preference Optimization (DPO); and (3) an Agentic Evaluation module containing a cross-family jury that emulates human critical judgment for unbiased, independent post-hoc assessment. Extensive experiments demonstrate that \textsc{PlotTwist} consistently outperforms all baselines, including frontier models, across multiple Narrative Quality Dimensions (NQDs), achieving higher win rates against all but the strongest baseline, with which it remains competitive. Further validation confirms strong sensitivity to narrative quality, as the framework reliably distinguishes plots derived from critically acclaimed versus widely panned screenplays. Together, these results establish structured, preference-based alignment as a resource-efficient approach to high-quality creative plot generation."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="thorat-etal-2026-plottwist">
<titleInfo>
<title>PlotTwist: A Creative Plot Generation Framework with Small Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Abhinav</namePart>
<namePart type="family">Thorat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ravi</namePart>
<namePart type="given">Kumar</namePart>
<namePart type="family">Kolla</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jyotin</namePart>
<namePart type="family">Goel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Madhav</namePart>
<namePart type="family">Kataria</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Niranjan</namePart>
<namePart type="family">Pedanekar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 3rd Workshop on Language Understanding in the Human-Machine Era (LUHME 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Rui</namePart>
<namePart type="family">Sousa-Silva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Henrique</namePart>
<namePart type="family">Lopes Cardoso</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maarit</namePart>
<namePart type="family">Koponen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antonio</namePart>
<namePart type="family">Pareja-Lora</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-471-2</identifier>
</relatedItem>
<abstract>Creative plot generation presents a fundamental challenge for language models: transforming a concise premise into a coherent narrative while satisfying multiple narrative constraints, including global coherence, character development, pacing, tone consistency, and emotional progression. Although recent Large Language Models (LLMs) demonstrate strong fluency on general-purpose tasks, they require preference alignment to perform well on domain-specific tasks such as creative plot generation. However, conducting such alignment at the scale of frontier LLMs is computationally prohibitive, significantly limiting accessibility and practical deployment. To address this, we present PlotTwist, a structured framework that enables Small Language Models (SLMs) with łeq 3B active parameters to generate high-quality, premise-conditioned plots competitive with frontier systems of vastly greater parameter scale. Our approach decomposes generation into three specialized components: (1) an Aspect Rating Reward Model, trained via a novel Positive–Negative prompting strategy; (2) a Mixture-of-Experts (MoE) plot generator aligned via Direct Preference Optimization (DPO); and (3) an Agentic Evaluation module containing a cross-family jury that emulates human critical judgment for unbiased, independent post-hoc assessment. Extensive experiments demonstrate that PlotTwist consistently outperforms all baselines, including frontier models, across multiple Narrative Quality Dimensions (NQDs), achieving higher win rates against all but the strongest baseline, with which it remains competitive. Further validation confirms strong sensitivity to narrative quality, as the framework reliably distinguishes plots derived from critically acclaimed versus widely panned screenplays. Together, these results establish structured, preference-based alignment as a resource-efficient approach to high-quality creative plot generation.</abstract>
<identifier type="citekey">thorat-etal-2026-plottwist</identifier>
<location>
<url>https://aclanthology.org/2026.luhme-1.13/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>144</start>
<end>172</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T PlotTwist: A Creative Plot Generation Framework with Small Language Models
%A Thorat, Abhinav
%A Kolla, Ravi Kumar
%A Goel, Jyotin
%A Kataria, Madhav
%A Pedanekar, Niranjan
%Y Sousa-Silva, Rui
%Y Lopes Cardoso, Henrique
%Y Koponen, Maarit
%Y Pareja-Lora, Antonio
%S Proceedings of the 3rd Workshop on Language Understanding in the Human-Machine Era (LUHME 2026)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-471-2
%F thorat-etal-2026-plottwist
%X Creative plot generation presents a fundamental challenge for language models: transforming a concise premise into a coherent narrative while satisfying multiple narrative constraints, including global coherence, character development, pacing, tone consistency, and emotional progression. Although recent Large Language Models (LLMs) demonstrate strong fluency on general-purpose tasks, they require preference alignment to perform well on domain-specific tasks such as creative plot generation. However, conducting such alignment at the scale of frontier LLMs is computationally prohibitive, significantly limiting accessibility and practical deployment. To address this, we present PlotTwist, a structured framework that enables Small Language Models (SLMs) with łeq 3B active parameters to generate high-quality, premise-conditioned plots competitive with frontier systems of vastly greater parameter scale. Our approach decomposes generation into three specialized components: (1) an Aspect Rating Reward Model, trained via a novel Positive–Negative prompting strategy; (2) a Mixture-of-Experts (MoE) plot generator aligned via Direct Preference Optimization (DPO); and (3) an Agentic Evaluation module containing a cross-family jury that emulates human critical judgment for unbiased, independent post-hoc assessment. Extensive experiments demonstrate that PlotTwist consistently outperforms all baselines, including frontier models, across multiple Narrative Quality Dimensions (NQDs), achieving higher win rates against all but the strongest baseline, with which it remains competitive. Further validation confirms strong sensitivity to narrative quality, as the framework reliably distinguishes plots derived from critically acclaimed versus widely panned screenplays. Together, these results establish structured, preference-based alignment as a resource-efficient approach to high-quality creative plot generation.
%U https://aclanthology.org/2026.luhme-1.13/
%P 144-172
Markdown (Informal)
[PlotTwist: A Creative Plot Generation Framework with Small Language Models](https://aclanthology.org/2026.luhme-1.13/) (Thorat et al., LUHME 2026)
ACL