# Meta Module Network for Compositional Visual Reasoning

arXiv: https://arxiv.org/abs/1910.03230 · HTML: https://mlsanity.com/paper/1910.03230/

## Key facts

- Cited by 39 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Oct 2019.

## Built on by (most cited first, 30 of 39)

- [MDETR -- Modulated Detection for End-to-End Multi-Modal Understanding](https://mlsanity.com/paper/2104.12763/) (2021)
- [Vision-Language Pre-training: Basics, Recent Advances, and Future Trends](https://mlsanity.com/paper/2210.09263/) (2022)
- [VinVL: Revisiting Visual Representations in Vision-Language Models](https://mlsanity.com/paper/2101.00529/) (2021)
- [NS3D: Neuro-Symbolic Grounding of 3D Objects and Relations](https://mlsanity.com/paper/2303.13483/) (2023)
- [See, Think, Confirm: Interactive Prompting Between Vision and Language Models for Knowledge-based Visual Reasoning](https://mlsanity.com/paper/2301.05226/) (2023)
- [3D Question Answering](https://mlsanity.com/paper/2112.08359/) (2021)
- [What's Left? Concept Grounding with Logic-Enhanced Foundation Models](https://mlsanity.com/paper/2310.16035/) (2023)
- [Explain Before You Answer: A Survey on Compositional Visual Reasoning](https://mlsanity.com/paper/2508.17298/) (2025)
- [Coarse-to-Fine Reasoning for Visual Question Answering](https://mlsanity.com/paper/2110.02526/) (2021)
- [ProTo: Program-Guided Transformer for Program-Guided Tasks](https://mlsanity.com/paper/2110.00804/) (2021)
- [SemVLP: Vision-Language Pre-training by Aligning Semantics at Multiple Levels](https://mlsanity.com/paper/2103.07829/) (2021)
- [GENOME: GenerativE Neuro-symbOlic visual reasoning by growing and reusing ModulEs](https://mlsanity.com/paper/2311.04901/) (2023)
- [How Transferable are Reasoning Patterns in VQA?](https://mlsanity.com/paper/2104.03656/) (2021)
- [Measuring Compositional Consistency for Video Question Answering](https://mlsanity.com/paper/2204.07190/) (2022)
- [Calibrating Concepts and Operations: Towards Symbolic Reasoning on Real Images](https://mlsanity.com/paper/2110.00519/) (2021)
- [Graphhopper: Multi-Hop Scene Graph Reasoning for Visual Question Answering](https://mlsanity.com/paper/2107.06325/) (2021)
- [Weakly Supervised Relative Spatial Reasoning for Visual Question Answering](https://mlsanity.com/paper/2109.01934/) (2021)
- [X-GGM: Graph Generative Modeling for Out-of-Distribution Generalization in Visual Question Answering](https://mlsanity.com/paper/2107.11576/) (2021)
- [Declaration-based Prompt Tuning for Visual Question Answering](https://mlsanity.com/paper/2205.02456/) (2022)
- [Uncovering the Full Potential of Visual Grounding Methods in VQA](https://mlsanity.com/paper/2401.07803/) (2024)
- [Measuring Faithful and Plausible Visual Grounding in VQA](https://mlsanity.com/paper/2305.15015/) (2023)
- [Supervising the Transfer of Reasoning Patterns in VQA](https://mlsanity.com/paper/2106.05597/) (2021)
- [MLP Architectures for Vision-and-Language Modeling: An Empirical Study](https://mlsanity.com/paper/2112.04453/) (2021)
- [Transformer Module Networks for Systematic Generalization in Visual Question Answering](https://mlsanity.com/paper/2201.11316/) (2022)
- [On the Role of Visual Grounding in VQA](https://mlsanity.com/paper/2406.18253/) (2024)
- [IMoRe: Implicit Program-Guided Reasoning for Human Motion Q&A](https://mlsanity.com/paper/2508.01984/) (2025)
- [From Shallow to Deep: Compositional Reasoning over Graphs for Visual Question Answering](https://mlsanity.com/paper/2206.12533/) (2022)
- [Visually Grounded VQA by Lattice-based Retrieval](https://mlsanity.com/paper/2211.08086/) (2022)
- [Modularized Zero-shot VQA with Pre-trained Models](https://mlsanity.com/paper/2305.17369/) (2023)
- [Detection-based Intermediate Supervision for Visual Question Answering](https://mlsanity.com/paper/2312.16012/) (2023)

## Abstract (arXiv)

Neural Module Network (NMN) exhibits strong interpretability and compositionality thanks to its handcrafted neural modules with explicit multi-hop reasoning capability. However, most NMNs suffer from two critical drawbacks: 1) scalability: customized module for specific function renders it impractical when scaling up to a larger set of functions in complex tasks; 2) generalizability: rigid pre-defined module inventory makes it difficult to generalize to unseen functions in new tasks/domains. To design a more powerful NMN architecture for practical use, we propose Meta Module Network (MMN) centered on a novel meta module, which can take in function recipes and morph into diverse instance modules dynamically. The instance modules are then woven into an execution graph for complex visual reasoning, inheriting the strong explainability and compositionality of NMN. With such a flexible instantiation mechanism, the parameters of instance modules are inherited from the central meta module, retaining the same model complexity as the function set grows, which promises better scalability. Meanwhile, as functions are encoded into the embedding space, unseen functions can be readily represented based on its structural similarity with previously observed ones, which ensures better generalizability. Experiments on GQA and CLEVR datasets validate the superiority of MMN over state-of-the-art NMN designs. Synthetic experiments on held-out unseen functions from GQA dataset also demonstrate the strong generalizability of MMN. Our code and model are released in Github https://github.com/wenhuchen/Meta-Module-Network.