# COLUMBUS: Evaluating COgnitive Lateral Understanding through Multiple-choice reBUSes

arXiv: https://arxiv.org/abs/2409.04053 · HTML: https://mlsanity.com/paper/2409.04053/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 41 references that are in the map.
- First posted on arXiv 6 Sep 2024.

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

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [OPT: Open Pre-trained Transformer Language Models](https://mlsanity.com/paper/2205.01068/) (2022)
- [Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond](https://mlsanity.com/paper/2308.12966/) (2023)
- [MMBench: Is Your Multi-modal Model an All-around Player?](https://mlsanity.com/paper/2307.06281/) (2023)
- [MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts](https://mlsanity.com/paper/2310.02255/) (2023)
- [MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers](https://mlsanity.com/paper/2002.10957/) (2020)
- [CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning](https://mlsanity.com/paper/1612.06890/) (2016)
- [RAVEN: A Dataset for Relational and Analogical Visual rEasoNing](https://mlsanity.com/paper/1903.02741/) (2019)
- [Measuring abstract reasoning in neural networks](https://mlsanity.com/paper/1807.04225/) (2018)
- [A Multi-World Approach to Question Answering about Real-World Scenes based on Uncertain Input](https://mlsanity.com/paper/1410.0210/) (2014)
- [Super-CLEVR: A Virtual Benchmark to Diagnose Domain Robustness in Visual Reasoning](https://mlsanity.com/paper/2212.00259/) (2022)
- [Breaking Common Sense: WHOOPS! A Vision-and-Language Benchmark of Synthetic and Compositional Images](https://mlsanity.com/paper/2303.07274/) (2023)
- [Bongard-LOGO: A New Benchmark for Human-Level Concept Learning and Reasoning](https://mlsanity.com/paper/2010.00763/) (2020)
- [Learning Representations that Support Extrapolation](https://mlsanity.com/paper/2007.05059/) (2020)
- [A Review of Emerging Research Directions in Abstract Visual Reasoning](https://mlsanity.com/paper/2202.10284/) (2022)
- [MARVEL: Multidimensional Abstraction and Reasoning through Visual Evaluation and Learning](https://mlsanity.com/paper/2404.13591/) (2024)
- [CANDLE: Iterative Conceptualization and Instantiation Distillation from Large Language Models for Commonsense Reasoning](https://mlsanity.com/paper/2401.07286/) (2024)
- [Machine Number Sense: A Dataset of Visual Arithmetic Problems for Abstract and Relational Reasoning](https://mlsanity.com/paper/2004.12193/) (2020)
- [SemEval-2024 Task 9: BRAINTEASER: A Novel Task Defying Common Sense](https://mlsanity.com/paper/2404.16068/) (2024)

## Built on by (most cited first, 5 of 5)

- [Puzzled by Puzzles: When Vision-Language Models Can't Take a Hint](https://mlsanity.com/paper/2505.23759/) (2025)
- [Reasoning or Pattern Matching? Probing Large Vision-Language Models with Visual Puzzles](https://mlsanity.com/paper/2601.13705/) (2026)
- [AssoCiAm: A Benchmark for Evaluating Association Thinking while Circumventing Ambiguity](https://mlsanity.com/paper/2509.14171/) (2025)
- [What's in Common? Multimodal Models Hallucinate When Reasoning Across Scenes](https://mlsanity.com/paper/2511.03768/) (2025)
- [CresOWLve: Benchmarking Creative Problem-Solving Over Real-World Knowledge](https://mlsanity.com/paper/2604.03374/) (2026)

## Abstract (arXiv)

While visual question-answering (VQA) benchmarks have catalyzed the development of reasoning techniques, they have focused on vertical thinking. Effective problem-solving also necessitates lateral thinking, which remains understudied in AI and has not been used to test visual perception systems. To bridge this gap, we formulate visual lateral thinking as a multiple-choice question-answering task and describe a three-step taxonomy-driven methodology for instantiating task examples. Then, we develop COLUMBUS, a synthetic benchmark that applies the task pipeline to create QA sets with text and icon rebus puzzles based on publicly available collections of compounds and common phrases. COLUMBUS comprises over 1,000 puzzles, each with four answer candidates. While the SotA vision-language models (VLMs) achieve decent performance, our evaluation demonstrates a substantial gap between humans and models. VLMs benefit from human-curated descriptions but struggle to self-generate such representations at the right level of abstraction.