COGITAO: A Visual Reasoning Framework To Study Compositionality & Generalization

• URL: http://arxiv.org/abs/2509.05249v1
• Date: Fri, 05 Sep 2025 17:01:05 GMT
• Title: COGITAO: A Visual Reasoning Framework To Study Compositionality & Generalization
• Authors: Yassine Taoudi-Benchekroun, Klim Troyan, Pascal Sager, Stefan Gerber, Lukas Tuggener, Benjamin Grewe,
• Abstract summary: COGITAO is a framework to study composition and generalization in visual domains.<n>It constructs rule-based tasks which apply a set of transformations to objects in grid-like environments.<n>It supports composition, at adjustable depth, over a set of 28 transformations, along with extensive control over grid parametrization and object properties.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The ability to compose learned concepts and apply them in novel settings is key to human intelligence, but remains a persistent limitation in state-of-the-art machine learning models. To address this issue, we introduce COGITAO, a modular and extensible data generation framework and benchmark designed to systematically study compositionality and generalization in visual domains. Drawing inspiration from ARC-AGI's problem-setting, COGITAO constructs rule-based tasks which apply a set of transformations to objects in grid-like environments. It supports composition, at adjustable depth, over a set of 28 interoperable transformations, along with extensive control over grid parametrization and object properties. This flexibility enables the creation of millions of unique task rules -- surpassing concurrent datasets by several orders of magnitude -- across a wide range of difficulties, while allowing virtually unlimited sample generation per rule. We provide baseline experiments using state-of-the-art vision models, highlighting their consistent failures to generalize to novel combinations of familiar elements, despite strong in-domain performance. COGITAO is fully open-sourced, including all code and datasets, to support continued research in this field.

Related papers

• OFA-MAS: One-for-All Multi-Agent System Topology Design based on Mixture-of-Experts Graph Generative Models [57.94189874119267]
  Multi-Agent Systems (MAS) offer a powerful paradigm for solving complex problems.<n>Current graph learning-based design methodologies often adhere to a "one-for-one" paradigm.<n>We propose OFA-TAD, a one-for-all framework that generates adaptive collaboration graphs for any task described in natural language.
  arXiv  Detail & Related papers  (2026-01-19T12:23:44Z)
• Domain-Specific Data Generation Framework for RAG Adaptation [58.20906914537952]
  Retrieval-Augmented Generation (RAG) combines the language understanding and reasoning power of large language models with external retrieval to enable domain-grounded responses.<n>We propose RAGen, a framework for generating domain-grounded question-answer-context (QAC) triples tailored to diverse RAG adaptation approaches.
  arXiv  Detail & Related papers  (2025-10-13T09:59:49Z)
• RAG-IGBench: Innovative Evaluation for RAG-based Interleaved Generation in Open-domain Question Answering [50.42577862494645]
  We present RAG-IGBench, a benchmark designed to evaluate the task of Interleaved Generation based on Retrieval-Augmented Generation (RAG-IG) in open-domain question answering.<n>RAG-IG integrates multimodal large language models (MLLMs) with retrieval mechanisms, enabling the models to access external image-text information for generating coherent multimodal content.
  arXiv  Detail & Related papers  (2025-10-11T03:06:39Z)
• SCoGen: Scenario-Centric Graph-Based Synthesis of Real-World Code Problems [6.798217798359233]
  We propose a novel framework for synthesizing code problems that emulate authentic real-world scenarios.<n>This framework systematically integrates domain knowledge, domain skills, and coding skills, all of which are extracted from real-world programming-related datasets.<n>The proposed method consistently achieves superior performance over state-of-the-art open-source large language models.
  arXiv  Detail & Related papers  (2025-09-16T12:52:48Z)
• Assemble Your Crew: Automatic Multi-agent Communication Topology Design via Autoregressive Graph Generation [72.44384066166147]
  Multi-agent systems (MAS) based on large language models (LLMs) have emerged as a powerful solution for dealing with complex problems across diverse domains.<n>Existing approaches are fundamentally constrained by their reliance on a template graph modification paradigm with a predefined set of agents and hard-coded interaction structures.<n>We propose ARG-Designer, a novel autoregressive model that operationalizes this paradigm by constructing the collaboration graph from scratch.
  arXiv  Detail & Related papers  (2025-07-24T09:17:41Z)
• Aggregated Structural Representation with Large Language Models for Human-Centric Layout Generation [7.980497203230983]
  We propose an Aggregation Structural Representation (ASR) module that integrates graph networks with large language models (LLMs) to preserve structural information while enhancing generative capability.<n>A comprehensive evaluation on the RICO dataset demonstrates the strong performance of ASR, both quantitatively using mean Intersection over Union (mIoU) and qualitatively through a crowdsourced user study.
  arXiv  Detail & Related papers  (2025-05-26T06:17:21Z)
• PuzzleBench: A Fully Dynamic Evaluation Framework for Large Multimodal Models on Puzzle Solving [50.50405233978406]
  We propose a fully dynamic multimodal evaluation framework, named Open-ended Visual Puzzle Generation (OVPG)<n>OVPG aims to generate fresh, diverse, and verifiable evaluation data automatically in puzzle-solving tasks.<n>Built upon OVPG, we construct PuzzleBench, a dynamic and scalable benchmark comprising 11,840 VQA samples.
  arXiv  Detail & Related papers  (2025-04-15T05:29:31Z)
• Adapting to Non-Stationary Environments: Multi-Armed Bandit Enhanced Retrieval-Augmented Generation on Knowledge Graphs [23.357843519762483]
  Recent studies have demonstrated that leveraging the Retrieval-Augmented Generation framework, combined with Knowledge Graphs, robustly enhances the reasoning capabilities of Large language models.<n>We introduce a Multi-objective Multi-Armed Bandit enhanced RAG framework, supported by multiple retrieval methods with diverse capabilities.<n>Our method significantly outperforms baseline methods in non-stationary settings while achieving state-of-the-art performance in stationary environments.
  arXiv  Detail & Related papers  (2024-12-10T15:56:03Z)
• Flex: End-to-End Text-Instructed Visual Navigation from Foundation Model Features [59.892436892964376]
  We investigate the minimal data requirements and architectural adaptations necessary to achieve robust closed-loop performance with vision-based control policies.<n>Our findings are synthesized in Flex (Fly lexically), a framework that uses pre-trained Vision Language Models (VLMs) as frozen patch-wise feature extractors.<n>We demonstrate the effectiveness of this approach on a quadrotor fly-to-target task, where agents trained via behavior cloning successfully generalize to real-world scenes.
  arXiv  Detail & Related papers  (2024-10-16T19:59:31Z)
• PosterLLaVa: Constructing a Unified Multi-modal Layout Generator with LLM [58.67882997399021]
  Our research introduces a unified framework for automated graphic layout generation.<n>Our data-driven method employs structured text (JSON format) and visual instruction tuning to generate layouts.<n>We develop an automated text-to-poster system that generates editable posters based on users' design intentions.
  arXiv  Detail & Related papers  (2024-06-05T03:05:52Z)
• Neural Constraint Satisfaction: Hierarchical Abstraction for Combinatorial Generalization in Object Rearrangement [75.9289887536165]
  We present a hierarchical abstraction approach to uncover underlying entities. We show how to learn a correspondence between intervening on states of entities in the agent's model and acting on objects in the environment. We use this correspondence to develop a method for control that generalizes to different numbers and configurations of objects.
  arXiv  Detail & Related papers  (2023-03-20T18:19:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.