A zero-shot table-text QA method that builds a hybrid graph of table cells and passage entities, prunes it via question-entity matching and 3-hop BFS, and feeds the pruned context to an LLM, improving EM and reducing tokens over full-context baselines.
Title resolution pending
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
-
Hybrid Graphs for Table-and-Text based Question Answering using LLMs
A zero-shot table-text QA method that builds a hybrid graph of table cells and passage entities, prunes it via question-entity matching and 3-hop BFS, and feeds the pruned context to an LLM, improving EM and reducing tokens over full-context baselines.