2026-06-04原文2 min readUpdated: 2026-06-04

POLARIS: Guiding Small Models to Write Long Stories

POLARIS is a training recipe for small open-weight models that significantly improves long-form creative writing. By combining a frontier LLM judge with human-reference injection in a GRPO framework, the resulting 9B model matches the performance of much larger models and exhibits strong length generalization.

SourcearXiv Computational LinguisticsAuthor: Rishanth Rajendhran, Jenna Russell, Mohit Iyyer, John Frederick Wieting

[2606.04095] POLARIS: Guiding Small Models to Write Long Stories

[Submitted on 2 Jun 2026]

Title:POLARIS: Guiding Small Models to Write Long Stories

View a PDF of the paper titled POLARIS: Guiding Small Models to Write Long Stories, by Rishanth Rajendhran and 3 other authors

View PDF

Abstract:Small open-weight models struggle at long-form creative writing: their generated stories either fall far short of the requested length, or their quality significantly degrades as length increases, especially when compared to frontier models. We present POLARIS (Policy Optimization with LLM-as-a-judge rewards and Anchored-Reference Injection for Storywriting), a lower-compute GRPO recipe with two key ingredients: a frontier LLM judge with a structured Story Quality rubric as the online reward, and human-reference injection (HRI), where a teacher-forced human-written story serves as a high-reward anchor within each GRPO group. By applying our training recipe to Qwen3.5-9B, using a dataset of approximately 1.4K prompt-story pairs derived from 100 short-story anthologies and 4 A100 GPUs, we obtain POLARIS-9B. Across five benchmarks spanning in-distribution and out-of-distribution prompts and rubrics, POLARIS-9B is competitive with much larger open-weight models while following length instructions more closely. A blinded human evaluation confirms that POLARIS-9B is preferred to the base Qwen3.5-9B and on par with Qwen3.5-27B. Despite training only on stories up to 4k words, POLARIS-9B preserves quality on prompts requesting stories up to 3 times the training length, a regime where most open-weight models degrade substantially in quality, length adherence, or both. More broadly, our results suggest that length generalization is a meaningful stress test for creative-writing models and a useful lens for distinguishing otherwise close models.

Subjects:

Computation and Language (cs.CL); Artificial Intelligence (cs.AI)

Cite as: arXiv:2606.04095 [cs.CL]

(or arXiv:2606.04095v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2606.04095

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Rishanth Rajendhran [view email] [v1] Tue, 2 Jun 2026 18:00:07 UTC (7,952 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled POLARIS: Guiding Small Models to Write Long Stories, by Rishanth Rajendhran and 3 other authors

View PDF

TeX Source

view license

Current browse context:

cs.CL

new | recent | 2026-06

Change to browse by:

cs cs.AI

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)