Table-Text Alignment: Explaining Claim Verification Against Tables in Scientific Papers
This addresses the need for interpretability in scientific claim verification, though it is incremental as it builds on existing benchmarks.
The paper tackles the problem of verifying scientific claims against tables by reframing it as an explanation task to identify essential table cells, and shows that incorporating alignment information improves claim verification performance, with LLMs often failing to recover human-aligned rationales.
Scientific claim verification against tables typically requires predicting whether a claim is supported or refuted given a table. However, we argue that predicting the final label alone is insufficient: it reveals little about the model's reasoning and offers limited interpretability. To address this, we reframe table-text alignment as an explanation task, requiring models to identify the table cells essential for claim verification. We build a new dataset by extending the SciTab benchmark with human-annotated cell-level rationales. Annotators verify the claim label and highlight the minimal set of cells needed to support their decision. After the annotation process, we utilize the collected information and propose a taxonomy for handling ambiguous cases. Our experiments show that (i) incorporating table alignment information improves claim verification performance, and (ii) most LLMs, while often predicting correct labels, fail to recover human-aligned rationales, suggesting that their predictions do not stem from faithful reasoning.