GAAMA 2.0: An Integrated System that Answers Boolean and Extractive Questions
This addresses the need for a unified system to handle multiple question types in multilingual QA, though it appears incremental as it builds on existing transformer-based approaches.
The paper tackles the problem of integrated support for answering both boolean and extractive questions in machine reading comprehension, presenting GAAMA 2.0, which achieves first place on the Tydi QA leaderboard.
Recent machine reading comprehension datasets include extractive and boolean questions but current approaches do not offer integrated support for answering both question types. We present a multilingual machine reading comprehension system and front-end demo that handles boolean questions by providing both a YES/NO answer and highlighting supporting evidence, and handles extractive questions by highlighting the answer in the passage. Our system, GAAMA 2.0, is ranked first on the Tydi QA leaderboard at the time of this writing. We contrast two different implementations of our approach. The first includes several independent stacks of transformers allowing easy deployment of each component. The second is a single stack of transformers utilizing adapters to reduce GPU memory footprint in a resource-constrained environment.