# Igniting Creative Writing in Small Language Models: LLM-as-a-Judge versus Multi-Agent Refined Rewards

arXiv: https://arxiv.org/abs/2508.21476 · HTML: https://mlsanity.com/paper/2508.21476/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use.
- Builds on 27 of its 59 references that are in the map.
- First posted on arXiv 29 Aug 2025.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)
- LLM alignment / RLHF / preference optimisation / [Human values and AI feedback alignment](https://mlsanity.com/alignment/human-values-and-ai-feedback-alignment/)

## Builds on (most cited first, 27 of 27)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Learning to summarize from human feedback](https://mlsanity.com/paper/2009.01325/) (2020)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [Improving Factuality and Reasoning in Language Models through Multiagent Debate](https://mlsanity.com/paper/2305.14325/) (2023)
- [Multitask Prompted Training Enables Zero-Shot Task Generalization](https://mlsanity.com/paper/2110.08207/) (2021)
- [LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://mlsanity.com/paper/2403.13372/) (2024)
- [LIMA: Less Is More for Alignment](https://mlsanity.com/paper/2305.11206/) (2023)
- [Diffusion-LM Improves Controllable Text Generation](https://mlsanity.com/paper/2205.14217/) (2022)
- [Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate](https://mlsanity.com/paper/2305.19118/) (2023)
- [TinyStories: How Small Can Language Models Be and Still Speak Coherent English?](https://mlsanity.com/paper/2305.07759/) (2023)
- [Re3: Generating Longer Stories With Recursive Reprompting and Revision](https://mlsanity.com/paper/2210.06774/) (2022)
- [Art or Artifice? Large Language Models and the False Promise of Creativity](https://mlsanity.com/paper/2309.14556/) (2023)
- [FutureSightDrive: Thinking Visually with Spatio-Temporal CoT for Autonomous Driving](https://mlsanity.com/paper/2505.17685/) (2025)
- [Branch-Solve-Merge Improves Large Language Model Evaluation and Generation](https://mlsanity.com/paper/2310.15123/) (2023)
- [Survey of Cultural Awareness in Language Models: Text and Beyond](https://mlsanity.com/paper/2411.00860/) (2024)
- [A Confederacy of Models: a Comprehensive Evaluation of LLMs on Creative Writing](https://mlsanity.com/paper/2310.08433/) (2023)
- [CULTURE-GEN: Revealing Global Cultural Perception in Language Models through Natural Language Prompting](https://mlsanity.com/paper/2404.10199/) (2024)
- [On The Origin of Cultural Biases in Language Models: From Pre-training Data to Linguistic Phenomena](https://mlsanity.com/paper/2501.04662/) (2025)
- [Automated Creativity Evaluation for Large Language Models: A Reference-Based Approach](https://mlsanity.com/paper/2504.15784/) (2025)
- [NLG Evaluation Metrics Beyond Correlation Analysis: An Empirical Metric Preference Checklist](https://mlsanity.com/paper/2305.08566/) (2023)
- [Multi-Modal Multi-Behavior Sequential Recommendation with Conditional Diffusion-Based Feature Denoising](https://mlsanity.com/paper/2508.05352/) (2025)
- [Controllable Text Generation for Open-Domain Creativity and Fairness](https://mlsanity.com/paper/2209.12099/) (2022)

## Built on by (most cited first, 11 of 11)

- [DPWriter: Reinforcement Learning with Diverse Planning Branching for Creative Writing](https://mlsanity.com/paper/2601.09609/) (2026)
- [LLM Review: Enhancing Creative Writing via Blind Peer Review Feedback](https://mlsanity.com/paper/2601.08003/) (2026)
- [MM-StanceDet: Retrieval-Augmented Multi-modal Multi-agent Stance Detection](https://mlsanity.com/paper/2604.27934/) (2026)
- [Retrieval, Reward, and Training Protocols: What Matters in Training Search Agents?](https://mlsanity.com/paper/2605.27881/) (2026)
- [Retell, Reward, Repeat: Reinforcement Learning for Narrative Theory-Informed Story Retelling](https://mlsanity.com/paper/2601.17226/) (2026)
- [UniCreative: Unifying Long-form Logic and Short-form Sparkle via Reference-Free Reinforcement Learning](https://mlsanity.com/paper/2604.05517/) (2026)
- [Triviality Corrected Endogenous Reward](https://mlsanity.com/paper/2604.11522/) (2026)
- [Retrieval-Augmented Multimodal Model for Fake News Detection](https://mlsanity.com/paper/2604.18112/) (2026)
- [From Coarse to Fine: Benchmarking and Reward Modeling for Writing-Centric Generation Tasks](https://mlsanity.com/paper/2604.27453/) (2026)
- [POLARIS: Guiding Small Models to Write Long Stories](https://mlsanity.com/paper/2606.04095/) (2026)
- [AgentCAT: Simulating Computerized Adaptive Testing via Multi-Agent Large Language Models](https://mlsanity.com/paper/2606.21832/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) have demonstrated remarkable creative writing capabilities, yet their substantial computational demands hinder widespread use. Enhancing Small Language Models (SLMs) offers a promising alternative, but current methods like Supervised Fine-Tuning (SFT) struggle with novelty, and Reinforcement Learning from Human Feedback (RLHF) is costly. This paper explores two distinct AI-driven reward strategies within a Reinforcement Learning from AI Feedback (RLAIF) framework to ignite the creative writing of a 7B-parameter SLM, specifically for generating Chinese greetings. The first strategy employs a RM trained on high-quality preference data curated by a novel multi-agent rejection sampling framework designed for creative tasks. The second, more novel strategy utilizes a principle-guided LLM-as-a-Judge, whose reward function is optimized via an adversarial training scheme with a reflection mechanism, to directly provide reward signals. Comprehensive experiments reveal that while both approaches significantly enhance creative output over baselines, the principle-guided LLM-as-a-Judge demonstrably yields superior generation quality. Furthermore, it offers notable advantages in training efficiency and reduced dependency on human-annotated data, presenting a more scalable and effective path towards creative SLMs. Our automated evaluation methods also exhibit strong alignment with human judgments. Our code and data are publicly available at https://github.com/weixiaolong94-hub/Igniting-Creative-Writing-in-Small-Language-Models.