في العديد من مهام معالجة اللغة الطبيعية، تعد استرجاع مرور وإعادة التعريف بمرتبة المقطع الإجراءان الرئيسيان في إيجاد المعلومات ذات الصلة وتحديدها. بما أن كل من الإجراءين يسهمان في الأداء النهائي، فمن المهم تحسينها بشكل مشترك من أجل تحقيق تحسن متبادل. في هذه الورقة، نقترح نهج تدريب مشترك رواية لاسترجاع المقطع الكثيف وإعادة إطلاق Reranking. مساهمة رئيسية هي أننا نقدم تقطير List Norwise الديناميكي، حيث نقوم بتصميم نهج تدريبي موحد للأسرار لكل من المسترد و Re-Ranker. أثناء التقطير الديناميكي، يمكن تحسين المسترد و Re-Ranker بشكل متكامل وفقا لمعلومات بعضهم البعض. نقترح أيضا استراتيجية تكبير البيانات الهجينة لبناء مثيلات تدريب متنوعة لنهج تدريب ListWise. تظهر تجارب واسعة فعالية نهجنا على كل من بيانات MSMARCO والأسئلة الطبيعية. يتوفر الكود الخاص بنا في https://github.com/paddlepaddle/rocketqa.
In various natural language processing tasks, passage retrieval and passage re-ranking are two key procedures in finding and ranking relevant information. Since both the two procedures contribute to the final performance, it is important to jointly optimize them in order to achieve mutual improvement. In this paper, we propose a novel joint training approach for dense passage retrieval and passage reranking. A major contribution is that we introduce the dynamic listwise distillation, where we design a unified listwise training approach for both the retriever and the re-ranker. During the dynamic distillation, the retriever and the re-ranker can be adaptively improved according to each other's relevance information. We also propose a hybrid data augmentation strategy to construct diverse training instances for listwise training approach. Extensive experiments show the effectiveness of our approach on both MSMARCO and Natural Questions datasets. Our code is available at https://github.com/PaddlePaddle/RocketQA.
References used
https://aclanthology.org/
In open-domain question answering, dense passage retrieval has become a new paradigm to retrieve relevant passages for finding answers. Typically, the dual-encoder architecture is adopted to learn dense representations of questions and passages for s
We study multi-answer retrieval, an under-explored problem that requires retrieving passages to cover multiple distinct answers for a given question. This task requires joint modeling of retrieved passages, as models should not repeatedly retrieve pa
Passage retrieval and ranking is a key task in open-domain question answering and information retrieval. Current effective approaches mostly rely on pre-trained deep language model-based retrievers and rankers. These methods have been shown to effect
Dense retrieval has shown great success for passage ranking in English. However, its effectiveness for non-English languages remains unexplored due to limitation in training resources. In this work, we explore different transfer techniques for docume
In the context of neural passage retrieval, we study three promising techniques: synthetic data generation, negative sampling, and fusion. We systematically investigate how these techniques contribute to the performance of the retrieval system and ho