Where is this coming from? Making groundedness count in the evaluation of Document VQA models
This addresses a critical evaluation gap in Document VQA for researchers and practitioners, though it is incremental as it builds on existing benchmarks.
The paper tackles the problem that current Document VQA evaluation metrics fail to account for semantic and multimodal groundedness, leading to hallucinations and errors being treated equally with well-grounded outputs. It proposes a new evaluation methodology that incorporates groundedness, validated with human judgment, and shows it better indicates model robustness and rewards better-calibrated answers.
Document Visual Question Answering (VQA) models have evolved at an impressive rate over the past few years, coming close to or matching human performance on some benchmarks. We argue that common evaluation metrics used by popular benchmarks do not account for the semantic and multimodal groundedness of a model's outputs. As a result, hallucinations and major semantic errors are treated the same way as well-grounded outputs, and the evaluation scores do not reflect the reasoning capabilities of the model. In response, we propose a new evaluation methodology that accounts for the groundedness of predictions with regard to the semantic characteristics of the output as well as the multimodal placement of the output within the input document. Our proposed methodology is parameterized in such a way that users can configure the score according to their preferences. We validate our scoring methodology using human judgment and show its potential impact on existing popular leaderboards. Through extensive analyses, we demonstrate that our proposed method produces scores that are a better indicator of a model's robustness and tends to give higher rewards to better-calibrated answers.