Denoising Table-Text Retrieval for Open-Domain Question Answering
This work addresses retrieval challenges in open-domain QA for users handling tabular and textual data, representing an incremental advancement with specific technical enhancements.
The paper tackled the problem of false-positive labels and cross-table reasoning in table-text open-domain question answering by proposing DoTTeR, which uses a denoised training dataset and integrates table-level ranking information, resulting in significant improvements in retrieval recall and downstream QA tasks.
In table-text open-domain question answering, a retriever system retrieves relevant evidence from tables and text to answer questions. Previous studies in table-text open-domain question answering have two common challenges: firstly, their retrievers can be affected by false-positive labels in training datasets; secondly, they may struggle to provide appropriate evidence for questions that require reasoning across the table. To address these issues, we propose Denoised Table-Text Retriever (DoTTeR). Our approach involves utilizing a denoised training dataset with fewer false positive labels by discarding instances with lower question-relevance scores measured through a false positive detection model. Subsequently, we integrate table-level ranking information into the retriever to assist in finding evidence for questions that demand reasoning across the table. To encode this ranking information, we fine-tune a rank-aware column encoder to identify minimum and maximum values within a column. Experimental results demonstrate that DoTTeR significantly outperforms strong baselines on both retrieval recall and downstream QA tasks. Our code is available at https://github.com/deokhk/DoTTeR.