Reproduce “Open Question Answering over Tables and Text” 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
Wenhu Chen, Ming-Wei Chang, Eva Schlinger, William Wang, William W. Cohen
The authors introduce OTT-QA, an open-domain benchmark requiring retrieval across Wikipedia tables and text, and a system that links likely table–passage companions before search and reads retrieved blocks jointly. Its fusion retriever with cross-block reader scores 27.2 exact match and 31.5 F1 on test-best, versus 9.6 and 13.1 for iterative retrieval with single-block reading; results are specific to this benchmark.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.