N2N-GQA: Noise-to-Narrative for Graph-Based Table-Text Question Answering Using LLMs

TL;DR

N2N-GQA introduces a zero-shot, graph-based evidence organization method for hybrid table-text question answering, significantly improving multi-hop reasoning accuracy without task-specific training.

Contribution

It is the first framework to construct dynamic evidence graphs from noisy retrievals for open-domain hybrid QA, enhancing multi-hop reasoning capabilities.

Findings

19.9-point EM improvement over baselines on OTT-QA

Achieves 48.80 EM without task-specific training

Graph-based evidence curation rivals fine-tuned models

Abstract

Multi-hop question answering over hybrid table-text data requires retrieving and reasoning across multiple evidence pieces from large corpora, but standard Retrieval-Augmented Generation (RAG) pipelines process documents as flat ranked lists, causing retrieval noise to obscure reasoning chains. We introduce N2N-GQA. To our knowledge, it is the first zeroshot framework for open-domain hybrid table-text QA that constructs dynamic evidence graphs from noisy retrieval outputs. Our key insight is that multi-hop reasoning requires understanding relationships between evidence pieces: by modeling documents as graph nodes with semantic relationships as edges, we identify bridge documents connecting reasoning steps, a capability absent in list-based retrieval. On OTT-QA, graph-based evidence curation provides a 19.9-point EM improvement over strong baselines, demonstrating that organizing retrieval results as structured graphs is critical for multihop reasoning. N2N-GQA achieves 48.80 EM, matching finetuned retrieval models (CORE: 49.0 EM) and approaching heavily optimized systems (COS: 56.9 EM) without any task specific training. This establishes graph-structured evidence organization as essential for scalable, zero-shot multi-hop QA systems and demonstrates that simple, interpretable graph construction can rival sophisticated fine-tuned approaches.