Harnessing Adaptive Topology Representations for Zero-Shot Graph Question Answering

2025-08-08

AI Summary

Existing large multimodal models (LMMs) rely on fixed topology representation formats (TRFs) for zero-shot graph question answering (Graph QA), leading to inaccurate or redundant responses. Method: We propose the first adaptive TRF selection framework for zero-shot Graph QA, comprising: (i) constructing a diverse set of topology representations $F_{ZS}$; (ii) introducing Graph Response Efficiency (GRE) as a unified evaluation metric; and (iii) training a TRF router—conditioned on question semantics—to dynamically select optimal TRFs during inference. To support this, we systematically analyze TRF properties and curate the TRF Preference (TRFP) dataset. Results: Our method significantly improves both answer accuracy and response conciseness across seven algorithmic domains and two cross-domain downstream tasks, while maintaining strong generalization and computational efficiency.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Question Answering
• Knowledge Representation and Reasoning: Qualitative Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Large Multimodal Models (LMMs) have shown generalized zero-shot capabilities in diverse domain question-answering (QA) tasks, including graph QA that involves complex graph topologies. However, most current approaches use only a single type of graph representation, namely Topology Representation Form (TRF), such as prompt-unified text descriptions or style-fixed visual styles. Those "one-size-fits-all" approaches fail to consider the specific preferences of different models or tasks, often leading to incorrect or overly long responses. To address this, we first analyze the characteristics and weaknesses of existing TRFs, and then design a set of TRFs, denoted by $F_{ZS}$, tailored to zero-shot graph QA. We then introduce a new metric, Graph Response Efficiency (GRE), which measures the balance between the performance and the brevity in graph QA. Built on these, we develop the DynamicTRF framework, which aims to improve both the accuracy and conciseness of graph QA. To be specific, DynamicTRF first creates a TRF Preference (TRFP) dataset that ranks TRFs based on their GRE scores, to probe the question-specific TRF preferences. Then it trains a TRF router on the TRFP dataset, to adaptively assign the best TRF from $F_{ZS}$ for each question during the inference. Extensive experiments across 7 in-domain algorithmic graph QA tasks and 2 out-of-domain downstream tasks show that DynamicTRF significantly enhances the zero-shot graph QA of LMMs in terms of accuracy

Problem

• Improving zero-shot graph QA with adaptive topology representations
• Addressing inefficiency in graph QA responses via DynamicTRF framework
• Enhancing accuracy and conciseness in multimodal graph question answering

Innovation

• Adaptive topology representations for graph QA
• DynamicTRF framework improves accuracy and conciseness
• TRF router assigns best representation per question