# Language Models can Self-Lengthen to Generate Long Texts

arXiv: https://arxiv.org/abs/2410.23933 · HTML: https://mlsanity.com/paper/2410.23933/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- Builds on 32 of its 41 references that are in the map.
- First posted on arXiv 31 Oct 2024.

## Builds on (most cited first, 30 of 32)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [Self-Instruct: Aligning Language Models with Self-Generated Instructions](https://mlsanity.com/paper/2212.10560/) (2022)
- [Qwen2 Technical Report](https://mlsanity.com/paper/2407.10671/) (2024)
- [LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://mlsanity.com/paper/2403.13372/) (2024)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [ZeRO: Memory Optimizations Toward Training Trillion Parameter Models](https://mlsanity.com/paper/1910.02054/) (2019)
- [Hierarchical Neural Story Generation](https://mlsanity.com/paper/1805.04833/) (2018)
- [A Diversity-Promoting Objective Function for Neural Conversation Models](https://mlsanity.com/paper/1510.03055/) (2015)
- [A Survey on Knowledge Distillation of Large Language Models](https://mlsanity.com/paper/2402.13116/) (2024)
- [Plan-And-Write: Towards Better Automatic Storytelling](https://mlsanity.com/paper/1811.05701/) (2018)
- [Re3: Generating Longer Stories With Recursive Reprompting and Revision](https://mlsanity.com/paper/2210.06774/) (2022)
- [Rephrasing the Web: A Recipe for Compute and Data-Efficient Language Modeling](https://mlsanity.com/paper/2401.16380/) (2024)
- [Self-Alignment with Instruction Backtranslation](https://mlsanity.com/paper/2308.06259/) (2023)
- [LongWriter: Unleashing 10,000+ Word Generation from Long Context LLMs](https://mlsanity.com/paper/2408.07055/) (2024)
- [Strategies for Structuring Story Generation](https://mlsanity.com/paper/1902.01109/) (2019)
- [RecurrentGPT: Interactive Generation of (Arbitrarily) Long Text](https://mlsanity.com/paper/2305.13304/) (2023)
- [Self-play with Execution Feedback: Improving Instruction-following Capabilities of Large Language Models](https://mlsanity.com/paper/2406.13542/) (2024)
- [AlignBench: Benchmarking Chinese Alignment of Large Language Models](https://mlsanity.com/paper/2311.18743/) (2023)
- [HelloBench: Evaluating Long Text Generation Capabilities of Large Language Models](https://mlsanity.com/paper/2409.16191/) (2024)
- [Plan, Write, and Revise: an Interactive System for Open-Domain Story Generation](https://mlsanity.com/paper/1904.02357/) (2019)
- [Wordcraft: a Human-AI Collaborative Editor for Story Writing](https://mlsanity.com/paper/2107.07430/) (2021)
- [Towards Scalable Automated Alignment of LLMs: A Survey](https://mlsanity.com/paper/2406.01252/) (2024)
- [Suri: Multi-constraint Instruction Following for Long-form Text Generation](https://mlsanity.com/paper/2406.19371/) (2024)
- [The What, Why, and How of Context Length Extension Techniques in Large Language Models -- A Detailed Survey](https://mlsanity.com/paper/2401.07872/) (2024)
- [Summarize, Outline, and Elaborate: Long-Text Generation via Hierarchical Supervision from Extractive Summaries](https://mlsanity.com/paper/2010.07074/) (2020)
- [Towards a Unified View of Preference Learning for Large Language Models: A Survey](https://mlsanity.com/paper/2409.02795/) (2024)

## Built on by (most cited first, 24 of 24)

- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Scaling of Search and Learning: A Roadmap to Reproduce o1 from Reinforcement Learning Perspective](https://mlsanity.com/paper/2412.14135/) (2024)
- [OmniThink: Expanding Knowledge Boundaries in Machine Writing through Thinking](https://mlsanity.com/paper/2501.09751/) (2025)
- [LongEval: A Comprehensive Analysis of Long-Text Generation Through a Plan-based Paradigm](https://mlsanity.com/paper/2502.19103/) (2025)
- [Shifting Long-Context LLMs Research from Input to Output](https://mlsanity.com/paper/2503.04723/) (2025)
- [LongWriter-Zero: Mastering Ultra-Long Text Generation via Reinforcement Learning](https://mlsanity.com/paper/2506.18841/) (2025)
- [Writing-RL: Advancing Long-form Writing via Adaptive Curriculum Reinforcement Learning](https://mlsanity.com/paper/2506.05760/) (2025)
- [ACE-RL: Adaptive Constraint-Enhanced Reward for Long-form Generation Reinforcement Learning](https://mlsanity.com/paper/2509.04903/) (2025)
- [LongDPO: Unlock Better Long-form Generation Abilities for LLMs via Critique-augmented Stepwise Information](https://mlsanity.com/paper/2502.02095/) (2025)
- [Generalizing From Short to Long: Effective Data Synthesis for Long-Context Instruction Tuning](https://mlsanity.com/paper/2502.15592/) (2025)
- [LagKV: Lag-Relative Information of the KV Cache Tells Which Tokens Are Important](https://mlsanity.com/paper/2504.04704/) (2025)
- [SuperWriter: Reflection-Driven Long-Form Generation with Large Language Models](https://mlsanity.com/paper/2506.04180/) (2025)
- [On Stable Long-Form Generation: Benchmarking and Mitigating Length Volatility](https://mlsanity.com/paper/2605.01357/) (2026)
- [Lost-in-the-Middle in Long-Text Generation: Synthetic Dataset, Evaluation Framework, and Mitigation](https://mlsanity.com/paper/2503.06868/) (2025)
- [R2-Write: Reflection and Revision for Open-Ended Writing with Deep Reasoning](https://mlsanity.com/paper/2604.03004/) (2026)
- [LongWeave: A Long-Form Generation Benchmark Bridging Real-World Relevance and Verifiability](https://mlsanity.com/paper/2510.24345/) (2025)
- [Writer-R1: Enhancing Generative Writing in LLMs via Memory-augmented Replay Policy Optimization](https://mlsanity.com/paper/2603.15061/) (2026)
- [The PLLuM Instruction Corpus](https://mlsanity.com/paper/2511.17161/) (2025)
- [PodBench: A Comprehensive Benchmark for Instruction-Aware Audio-Oriented Podcast Script Generation](https://mlsanity.com/paper/2601.14903/) (2026)
- [LARFT: Closing the Cognition-Action Gap for Length Instruction Following in Large Language Models](https://mlsanity.com/paper/2603.19255/) (2026)
- [Measuring Information Distortion in Hierarchical Ultra long Novel Reconstruction:The Optimal Expansion Ratio](https://mlsanity.com/paper/2505.12572/) (2025)
- [HiCaM: A Hierarchical-Causal Modification Framework for Long-Form Text Modification](https://mlsanity.com/paper/2505.24319/) (2025)
- [From Coarse to Fine: Benchmarking and Reward Modeling for Writing-Centric Generation Tasks](https://mlsanity.com/paper/2604.27453/) (2026)
- [Scaling Long-Form Story Generation via Narrative State Tracking](https://mlsanity.com/paper/2609.35759/) (2026)

## Abstract (arXiv)

Recent advancements in Large Language Models (LLMs) have significantly enhanced their ability to process long contexts, yet a notable gap remains in generating long, aligned outputs. This limitation stems from a training gap where pre-training lacks effective instructions for long-text generation, and post-training data primarily consists of short query-response pairs. Current approaches, such as instruction backtranslation and behavior imitation, face challenges including data quality, copyright issues, and constraints on proprietary model usage. In this paper, we introduce an innovative iterative training framework called Self-Lengthen that leverages only the intrinsic knowledge and skills of LLMs without the need for auxiliary data or proprietary models. The framework consists of two roles: the Generator and the Extender. The Generator produces the initial response, which is then split and expanded by the Extender. This process results in a new, longer response, which is used to train both the Generator and the Extender iteratively. Through this process, the models are progressively trained to handle increasingly longer responses. Experiments on benchmarks and human evaluations show that Self-Lengthen outperforms existing methods in long-text generation, when applied to top open-source LLMs such as Qwen2 and LLaMA3. Our code is publicly available at https://github.com/QwenLM/Self-Lengthen.