解决大模型长文本生成中逻辑崩溃问题，让写作更连贯持久。
IS-CoT: Breaking the Long-form Generation Collapse via Interleaved Structural Thinking
- 采用动态计划-写作-反思循环，实时调整生成策略。
- 在超2000词长文上表现超越现有模型3.08点，接近闭源大模型。
- 适合需要长期一致性的内容创作场景，如报告、小说等。
长文本生成的连贯性与可控性仍是大语言模型的核心挑战。尽管增强推理的模型在逻辑密集任务中表现优异，但我们的评估显示，其在开放式写作中存在严重长度衰减问题——当目标长度超过2000词时性能急剧下降。我们归因于静态层次规划无法在长上下文中提供动态引导。为此，提出交织式结构思维（IS-CoT）框架，将动态的计划-写作-反思循环内嵌于生成过程，实现无需外部辅助的持续策略适应与全局对齐。基于此框架，通过多教师流水线构建高质量的交织推理数据集，并训练出IS-Writer-8B模型。实验表明，该模型在多个长文本基准测试中达到领先水平（如在LongBench-Write上较DeepSeek-V3.2提升3.08分），具备强长度合规性和与更大闭源模型相当的连贯性。
原文摘要 · Abstract (English)
Generating coherent and controllable long-form content remains a persistent challenge for Large Language Models (LLMs). While reasoning-enhanced models have demonstrated success in logic-intensive domains, our evaluation reveals that they suffer from a severe length collapse in open-ended writing, where performance degrades sharply as target lengths exceed 2,000 words. We attribute this failure to the limitation of static hierarchical planning, which struggles to provide dynamic guidance over extended contexts. To bridge this gap, we introduce the Interleaved Structural Chain-of-Thought (IS-CoT) framework. Unlike external agentic workflows, IS-CoT embeds a dynamic Plan-Write-Reflect cycle into the generation process, enabling continuous strategy adaptation and global alignment without additional assistance. Based on this framework, we construct a high-quality dataset of interleaved reasoning traces via a multi-teacher pipeline and train IS-Writer-8B. Experiments demonstrate that IS-Writer-8B achieves state-of-the-art performance on challenging long-form benchmarks (e.g., +3.08 vs. DeepSeek-V3.2 on LongBench-Write), exhibiting robust length compliance and coherence competitive with significantly larger proprietary models.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。