DyRRen: A Dynamic Retriever-Reranker-Generator Model for Numerical Reasoning over Tabular and Textual Data

TL;DR

DyRRen introduces a dynamic reranking mechanism in the retriever-reranker-generator framework to improve numerical reasoning over hybrid data, leading to better performance on the FinQA dataset.

Contribution

The paper presents DyRRen, a novel model that dynamically reranks retrieved sentences at each reasoning step, enhancing numerical reasoning over tables and texts.

Findings

Outperforms existing baselines on FinQA dataset

Demonstrates improved reasoning accuracy with dynamic reranking

Enhances retrieval relevance for each generation step

Abstract

Numerical reasoning over hybrid data containing tables and long texts has recently received research attention from the AI community. To generate an executable reasoning program consisting of math and table operations to answer a question, state-of-the-art methods use a retriever-generator pipeline. However, their retrieval results are static, while different generation steps may rely on different sentences. To attend to the retrieved information that is relevant to each generation step, in this paper, we propose DyRRen, an extended retriever-reranker-generator framework where each generation step is enhanced by a dynamic reranking of retrieved sentences. It outperforms existing baselines on the FinQA dataset.