RoR: Read-over-Read for Long Document Machine Reading Comprehension

TL;DR

RoR introduces a read-over-read approach that enhances long document comprehension by expanding the reading field, combining chunk and document-level reading with answer aggregation, significantly improving performance on benchmarks.

Contribution

The paper proposes RoR, a novel method that enables effective long document reading by integrating chunk-level and document-level comprehension with answer voting.

Findings

RoR achieves state-of-the-art results on QuAC and TriviaQA.

RoR ranks 1st on the QuAC leaderboard as of May 2021.

The method effectively handles long documents beyond traditional length limits.

Abstract

Transformer-based pre-trained models, such as BERT, have achieved remarkable results on machine reading comprehension. However, due to the constraint of encoding length (e.g., 512 WordPiece tokens), a long document is usually split into multiple chunks that are independently read. It results in the reading field being limited to individual chunks without information collaboration for long document machine reading comprehension. To address this problem, we propose RoR, a read-over-read method, which expands the reading field from chunk to document. Specifically, RoR includes a chunk reader and a document reader. The former first predicts a set of regional answers for each chunk, which are then compacted into a highly-condensed version of the original document, guaranteeing to be encoded once. The latter further predicts the global answers from this condensed document. Eventually, a voting strategy is utilized to aggregate and rerank the regional and global answers for final prediction. Extensive experiments on two benchmarks QuAC and TriviaQA demonstrate the effectiveness of RoR for long document reading. Notably, RoR ranks 1st place on the QuAC leaderboard (https://quac.ai/) at the time of submission (May 17th, 2021).