# MMAPG: A Training-Free Framework for Multimodal Multi-hop Question Answering via Adaptive Planning Graphs

arXiv: https://arxiv.org/abs/2508.16051 · HTML: https://mlsanity.com/paper/2508.16051/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 33 references that are in the map.
- First posted on arXiv 22 Aug 2025.

## Builds on (most cited first, 20 of 20)

- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022)
- [Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions](https://mlsanity.com/paper/2212.10509/) (2022)
- [Graph of Thoughts: Solving Elaborate Problems with Large Language Models](https://mlsanity.com/paper/2308.09687/) (2023)
- [Think-on-Graph: Deep and Responsible Reasoning of Large Language Model on Knowledge Graph](https://mlsanity.com/paper/2307.07697/) (2023)
- [MuRAG: Multimodal Retrieval-Augmented Generator for Open Question Answering over Images and Text](https://mlsanity.com/paper/2210.02928/) (2022)
- [ReWOO: Decoupling Reasoning from Observations for Efficient Augmented Language Models](https://mlsanity.com/paper/2305.18323/) (2023)
- [MultiModalQA: Complex Question Answering over Text, Tables and Images](https://mlsanity.com/paper/2104.06039/) (2021)
- [REVEAL: Retrieval-Augmented Visual-Language Pre-Training with Multi-Source Multimodal Knowledge Memory](https://mlsanity.com/paper/2212.05221/) (2022)
- [HybridRAG: Integrating Knowledge Graphs and Vector Retrieval Augmented Generation for Efficient Information Extraction](https://mlsanity.com/paper/2408.04948/) (2024)
- [Searching for Best Practices in Retrieval-Augmented Generation](https://mlsanity.com/paper/2407.01219/) (2024)
- [Paths-over-Graph: Knowledge Graph Empowered Large Language Model Reasoning](https://mlsanity.com/paper/2410.14211/) (2024)
- [Does CLIP's Generalization Performance Mainly Stem from High Train-Test Similarity?](https://mlsanity.com/paper/2310.09562/) (2023)
- [Enhancing Multi-modal and Multi-hop Question Answering via Structured Knowledge and Unified Retrieval-Generation](https://mlsanity.com/paper/2212.08632/) (2022)
- [Unified Language Representation for Question Answering over Text, Tables, and Images](https://mlsanity.com/paper/2306.16762/) (2023)
- [Chain of Thought with Explicit Evidence Reasoning for Few-shot Relation Extraction](https://mlsanity.com/paper/2311.05922/) (2023)
- [MMHQA-ICL: Multimodal In-context Learning for Hybrid Question Answering over Text, Tables and Images](https://mlsanity.com/paper/2309.04790/) (2023)
- [Semi-Structured Chain-of-Thought: Integrating Multiple Sources of Knowledge for Improved Language Model Reasoning](https://mlsanity.com/paper/2311.08505/) (2023)
- [Progressive Evidence Refinement for Open-domain Multimodal Retrieval Question Answering](https://mlsanity.com/paper/2310.09696/) (2023)
- [An Entailment Tree Generation Approach for Multimodal Multi-Hop Question Answering with Mixture-of-Experts and Iterative Feedback Mechanism](https://mlsanity.com/paper/2412.05821/) (2024)

## Built on by (most cited first, 1 of 1)

- [Failure is Feedback: History-Aware Backtracking for Agentic Traversal in Multimodal Graphs](https://mlsanity.com/paper/2602.03432/) (2026)

## Abstract (arXiv)

Multimodal Multi-hop question answering requires integrating information from diverse sources, such as images and texts, to derive answers. Existing methods typically rely on sequential retrieval and reasoning, where each step builds on the previous output. However, this single-path paradigm makes them vulnerable to errors due to misleading intermediate steps. Moreover, developing multimodal models can be computationally expensive, often requiring extensive training. To address these limitations, we propose a training-free framework guided by an Adaptive Planning Graph, which consists of planning, retrieval and reasoning modules. The planning module analyzes the current state of the Adaptive Planning Graph, determines the next action and where to expand the graph, which enables dynamic and flexible exploration of reasoning paths. To handle retrieval of text to unspecified target modalities, we devise modality-specific strategies that dynamically adapt to distinct data types. Our approach preserves the characteristics of multimodal information without costly task-specific training, enabling seamless integration with up-to-date models. Finally, the experiments on MultimodalQA and WebQA show that our approach matches or outperforms existing models that rely on training.