论文标题
实体条件的问题生成，用于神经信息检索中强大的注意力分布
Entity-Conditioned Question Generation for Robust Attention Distribution in Neural Information Retrieval
论文作者
论文摘要
我们表明，监督神经信息检索（IR）模型容易学习通行令牌上的稀疏注意模式，这可能导致关键短语，包括命名的实体受到较低的注意权重，最终导致模型不足。使用一种新颖的靶向合成数据生成方法，该方法识别出参与的实体和状况不佳的那些发作的发作，我们教会神经IR在给定段落中对所有实体更加均匀，更稳定。在两个公共IR基准测试中，我们从经验上表明，所提出的方法有助于改善模型的注意力模式和检索性能，包括在零弹位设置中。
We show that supervised neural information retrieval (IR) models are prone to learning sparse attention patterns over passage tokens, which can result in key phrases including named entities receiving low attention weights, eventually leading to model under-performance. Using a novel targeted synthetic data generation method that identifies poorly attended entities and conditions the generation episodes on those, we teach neural IR to attend more uniformly and robustly to all entities in a given passage. On two public IR benchmarks, we empirically show that the proposed method helps improve both the model's attention patterns and retrieval performance, including in zero-shot settings.
