ATF prunes table columns and rows with LLM scoring plus retrieval, cutting cells by about 70% and improving out-of-domain TableQA accuracy, while hurting in-domain QA and fact verification.
Binder: Binding language models in symbolic languages
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
What to Keep and What to Drop: Adaptive Table Filtering Framework
ATF prunes table columns and rows with LLM scoring plus retrieval, cutting cells by about 70% and improving out-of-domain TableQA accuracy, while hurting in-domain QA and fact verification.