Saved in:
Bibliographic Details
Main Authors: Vlachos, Angelos, Filandrianos, Giorgos, Lymperaiou, Maria, Spanos, Nikolaos, Mitsouras, Ilias, Karampinis, Vasileios, Voulodimos, Athanasios
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.00356
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916875075584000
author Vlachos, Angelos
Filandrianos, Giorgos
Lymperaiou, Maria
Spanos, Nikolaos
Mitsouras, Ilias
Karampinis, Vasileios
Voulodimos, Athanasios
author_facet Vlachos, Angelos
Filandrianos, Giorgos
Lymperaiou, Maria
Spanos, Nikolaos
Mitsouras, Ilias
Karampinis, Vasileios
Voulodimos, Athanasios
contents We present a Collaborative Agent-Based Framework for Multi-Image Reasoning. Our approach tackles the challenge of interleaved multimodal reasoning across diverse datasets and task formats by employing a dual-agent system: a language-based PromptEngineer, which generates context-aware, task-specific prompts, and a VisionReasoner, a large vision-language model (LVLM) responsible for final inference. The framework is fully automated, modular, and training-free, enabling generalization across classification, question answering, and free-form generation tasks involving one or multiple input images. We evaluate our method on 18 diverse datasets from the 2025 MIRAGE Challenge (Track A), covering a broad spectrum of visual reasoning tasks including document QA, visual comparison, dialogue-based understanding, and scene-level inference. Our results demonstrate that LVLMs can effectively reason over multiple images when guided by informative prompts. Notably, Claude 3.7 achieves near-ceiling performance on challenging tasks such as TQA (99.13% accuracy), DocVQA (96.87%), and MMCoQA (75.28 ROUGE-L). We also explore how design choices-such as model selection, shot count, and input length-influence the reasoning performance of different LVLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2508_00356
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Analyze-Prompt-Reason: A Collaborative Agent-Based Framework for Multi-Image Vision-Language Reasoning
Vlachos, Angelos
Filandrianos, Giorgos
Lymperaiou, Maria
Spanos, Nikolaos
Mitsouras, Ilias
Karampinis, Vasileios
Voulodimos, Athanasios
Computer Vision and Pattern Recognition
Multiagent Systems
I.2; I.2.7
We present a Collaborative Agent-Based Framework for Multi-Image Reasoning. Our approach tackles the challenge of interleaved multimodal reasoning across diverse datasets and task formats by employing a dual-agent system: a language-based PromptEngineer, which generates context-aware, task-specific prompts, and a VisionReasoner, a large vision-language model (LVLM) responsible for final inference. The framework is fully automated, modular, and training-free, enabling generalization across classification, question answering, and free-form generation tasks involving one or multiple input images. We evaluate our method on 18 diverse datasets from the 2025 MIRAGE Challenge (Track A), covering a broad spectrum of visual reasoning tasks including document QA, visual comparison, dialogue-based understanding, and scene-level inference. Our results demonstrate that LVLMs can effectively reason over multiple images when guided by informative prompts. Notably, Claude 3.7 achieves near-ceiling performance on challenging tasks such as TQA (99.13% accuracy), DocVQA (96.87%), and MMCoQA (75.28 ROUGE-L). We also explore how design choices-such as model selection, shot count, and input length-influence the reasoning performance of different LVLMs.
title Analyze-Prompt-Reason: A Collaborative Agent-Based Framework for Multi-Image Vision-Language Reasoning
topic Computer Vision and Pattern Recognition
Multiagent Systems
I.2; I.2.7
url https://arxiv.org/abs/2508.00356