让大模型学会反向理解文本，提升多语言理解能力。
Reverse Modeling in Large Language Models
- 用正反向文本联合训练，使模型具备双向理解能力
- 基于正反向损失差异筛选数据，显著提升模型性能
- 适合希望提升模型泛化与逆向推理能力的研究者
人类习惯正向阅读和写作，这种自然偏见也体现在自回归大语言模型对文本的理解中。本文研究大模型是否像人类一样难以处理反向输入文本。实验发现，公开预训练的大模型无法理解反向文本；而从头训练并同时使用正向与反向文本的模型，在推理时可跨多种语言实现同等理解能力。案例分析显示，不同内容的文本在正反方向输入时产生不同损失：部分文本正向损失更低，部分反向更低。据此提出一种基于正反向损失差值的数据选择方法。在持续预训练中使用该方法筛选的数据，能大幅提高大模型在多个语言理解基准上的表现。
原文摘要 · Abstract (English)
Humans are accustomed to reading and writing in a forward manner, and this natural bias extends to text understanding in auto-regressive large language models (LLMs). This paper investigates whether LLMs, like humans, struggle with reverse modeling, specifically with reversed text inputs. We found that publicly available pre-trained LLMs cannot understand such inputs. However, LLMs trained from scratch with both forward and reverse texts can understand them equally well during inference across multiple languages. Our case study shows that different-content texts result in different losses if input (to LLMs) in different directions -- some get lower losses for forward while some for reverse. This leads us to a simple and nice solution for data selection based on the loss differences between forward and reverse directions. Using our selected data in continued pretraining can boost LLMs' performance by a large margin across different language understanding benchmarks.
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。