Reproduce “Scientific Fact-Checking: A Survey of Resources and Approaches” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Juraj Vladika, Florian Matthes

This survey systematically reviews the landscape of automated scientific fact-checking, categorizing existing datasets and methodological approaches to verify scientific claims. It identifies core challenges and proposes future research directions to advance the field.

Key insights

  • Domain-specific language models (e.g., BioBERT, SciBERT) consistently demonstrate superior performance over general-purpose models for scientific claim verification.
  • Joint multi-task learning approaches, which integrate rationale selection and verdict prediction, generally outperform pipeline-based methods by preserving contextual information.
  • Existing datasets, while diverse in claim origin and evidence sources, often present limitations in size and dynamic representation of evolving scientific knowledge.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.