Saved in:
Bibliographic Details
Main Authors: Yuan, Yangshu, Chen, Heng, Jiang, Xinyi, Ng, Christian, Qiu, Kexin
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2507.22074
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916869242355712
author Yuan, Yangshu
Chen, Heng
Jiang, Xinyi
Ng, Christian
Qiu, Kexin
author_facet Yuan, Yangshu
Chen, Heng
Jiang, Xinyi
Ng, Christian
Qiu, Kexin
contents The rapid advancement of Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) has enhanced our ability to process and generate human language and visual information. However, these models often struggle with complex, multi-step multi-modal instructions that require logical reasoning, dynamic feedback integration, and iterative self-correction. To address this, we propose CIMR: Contextualized Iterative Multimodal Reasoning, a novel framework that introduces a context-aware iterative reasoning and self-correction module. CIMR operates in two stages: initial reasoning and response generation, followed by iterative refinement using parsed multi-modal feedback. A dynamic fusion module deeply integrates textual, visual, and contextual features at each step. We fine-tune LLaVA-1.5-7B on the Visual Instruction Tuning (VIT) dataset and evaluate CIMR on the newly introduced Multi-modal Action Planning (MAP) dataset. CIMR achieves 91.5% accuracy, outperforming state-of-the-art models such as GPT-4V (89.2%), LLaVA-1.5 (78.5%), MiniGPT-4 (75.3%), and InstructBLIP (72.8%), demonstrating the efficacy of its iterative reasoning and self-correction capabilities in complex tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2507_22074
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CIMR: Contextualized Iterative Multimodal Reasoning for Robust Instruction Following in LVLMs
Yuan, Yangshu
Chen, Heng
Jiang, Xinyi
Ng, Christian
Qiu, Kexin
Machine Learning
Computation and Language
The rapid advancement of Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) has enhanced our ability to process and generate human language and visual information. However, these models often struggle with complex, multi-step multi-modal instructions that require logical reasoning, dynamic feedback integration, and iterative self-correction. To address this, we propose CIMR: Contextualized Iterative Multimodal Reasoning, a novel framework that introduces a context-aware iterative reasoning and self-correction module. CIMR operates in two stages: initial reasoning and response generation, followed by iterative refinement using parsed multi-modal feedback. A dynamic fusion module deeply integrates textual, visual, and contextual features at each step. We fine-tune LLaVA-1.5-7B on the Visual Instruction Tuning (VIT) dataset and evaluate CIMR on the newly introduced Multi-modal Action Planning (MAP) dataset. CIMR achieves 91.5% accuracy, outperforming state-of-the-art models such as GPT-4V (89.2%), LLaVA-1.5 (78.5%), MiniGPT-4 (75.3%), and InstructBLIP (72.8%), demonstrating the efficacy of its iterative reasoning and self-correction capabilities in complex tasks.
title CIMR: Contextualized Iterative Multimodal Reasoning for Robust Instruction Following in LVLMs
topic Machine Learning
Computation and Language
url https://arxiv.org/abs/2507.22074