针对长文本生成中‘中间信息丢失’问题，提出新数据集与缓解方法。
Lost-in-the-Middle in Long-Text Generation: Synthetic Dataset, Evaluation Framework, and Mitigation
- 构建合成数据集与评估框架，填补长输入长输出任务的基准空白。
- 提出RAL-Writer模型，通过检索重述关键内容缓解中间信息丢失。
- 实验验证方法有效，适合长文本生成研究者参考使用。
现有长文本生成方法主要关注从短输入生成长文本，忽视了长输入长输出任务。这类任务具有广泛应用价值，但缺乏可用的基准。随着输入长度增加，现有方法不可避免地出现“中间信息丢失”现象。本文首先提出长输入长输出基准（LongInOutBench），包含合成数据集和全面评估框架，解决基准缺失问题。随后，提出检索增强型长文本生成器RAL-Writer，通过检索并重述被忽略的关键内容，构建显式提示以缓解“中间信息丢失”。最后，利用所提基准对RAL-Writer与基线方法进行对比评估，结果证明该方法有效。代码已开源于https://github.com/OnlyAR/RAL-Writer。
原文摘要 · Abstract (English)
Existing long-text generation methods primarily concentrate on producing lengthy texts from short inputs, neglecting the long-input and long-output tasks. Such tasks have numerous practical applications while lacking available benchmarks. Moreover, as the input grows in length, existing methods inevitably encounter the "lost-in-the-middle" phenomenon. In this paper, we first introduce a Long Input and Output Benchmark (LongInOutBench), including a synthetic dataset and a comprehensive evaluation framework, addressing the challenge of the missing benchmark. We then develop the Retrieval-Augmented Long-Text Writer (RAL-Writer), which retrieves and restates important yet overlooked content, mitigating the "lost-in-the-middle" issue by constructing explicit prompts. We finally employ the proposed LongInOutBench to evaluate our RAL-Writer against comparable baselines, and the results demonstrate the effectiveness of our approach. Our code has been released at https://github.com/OnlyAR/RAL-Writer.
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。