تقدم التطورات الحديثة في QA في الهواء الطلق إلى نماذج قوية تعتمد على استرجاع كثيف، ولكن ركزت فقط على استرداد المقاطع النصية.في هذا العمل، نتعامل مع QA المجال المفتوح على الجداول لأول مرة، وإظهار أنه يمكن تحسين الاسترجاع من خلال المسترد المصمم للتعامل مع سياق الجدول.نقدم إجراءات فعالة مسبقة التدريب لاستردادنا وتحسين جودة الاسترجاع مع السلبيات الصلبة الملغومة.نظرا لأن مجموعات البيانات ذات الصلة مفقودة، فإننا نستخلص مجموعة فرعية من الأسئلة الطبيعية (Kwiatkowski et al.، 2019) في مجموعة بيانات QA.نجد أن المسترد الخاص بنا يحسن نتائج الاسترجاع من 72.0 إلى 81.1 استدعاء @ 10 وتنفذ QA نهاية إلى نهاية من 33.8 إلى 37.7 مباراة دقيقة، عبر المسترد القائم على بيرت.
Recent advances in open-domain QA have led to strong models based on dense retrieval, but only focused on retrieving textual passages. In this work, we tackle open-domain QA over tables for the first time, and show that retrieval can be improved by a retriever designed to handle tabular context. We present an effective pre-training procedure for our retriever and improve retrieval quality with mined hard negatives. As relevant datasets are missing, we extract a subset of Natural Questions (Kwiatkowski et al., 2019) into a Table QA dataset. We find that our retriever improves retrieval results from 72.0 to 81.1 recall@10 and end-to-end QA results from 33.8 to 37.7 exact match, over a BERT based retriever.
References used
https://aclanthology.org/
Dense neural text retrieval has achieved promising results on open-domain Question Answering (QA), where latent representations of questions and passages are exploited for maximum inner product search in the retrieval process. However, current dense
We introduce SPARTA, a novel neural retrieval method that shows great promise in performance, generalization, and interpretability for open-domain question answering. Unlike many neural ranking methods that use dense vector nearest neighbor search, S
We introduce a new dataset for Question Rewriting in Conversational Context (QReCC), which contains 14K conversations with 80K question-answer pairs. The task in QReCC is to find answers to conversational questions within a collection of 10M web page
Multilingual question answering tasks typically assume that answers exist in the same language as the question. Yet in practice, many languages face both information scarcity---where languages have few reference articles---and information asymmetry--
Open-domain question answering aims at locating the answers to user-generated questions in massive collections of documents. Retriever-readers and knowledge graph approaches are two big families of solutions to this task. A retriever-reader first app