Abstract, Rationale, Stance: A Joint Model for Scientific Claim Verification
This work addresses scientific claim verification for researchers, offering an incremental improvement by enhancing joint learning and regularization in a known framework.
The paper tackles the problem of error propagation and lack of information sharing in pipeline models for scientific claim verification by proposing ARSJoint, a joint learning approach that integrates abstract retrieval, rationale selection, and stance prediction, which outperforms existing works on the SciFact benchmark dataset.
Scientific claim verification can help the researchers to easily find the target scientific papers with the sentence evidence from a large corpus for the given claim. Some existing works propose pipeline models on the three tasks of abstract retrieval, rationale selection and stance prediction. Such works have the problems of error propagation among the modules in the pipeline and lack of sharing valuable information among modules. We thus propose an approach, named as ARSJoint, that jointly learns the modules for the three tasks with a machine reading comprehension framework by including claim information. In addition, we enhance the information exchanges and constraints among tasks by proposing a regularization term between the sentence attention scores of abstract retrieval and the estimated outputs of rational selection. The experimental results on the benchmark dataset SciFact show that our approach outperforms the existing works.