paper-with-me

Papers

Decomposing Visual Classification: Assessing Tree-Based Reasoning in VLMs

2025-09-10 · Sary Elmansoury, Islam Mesabah, Gerrit Großmann, Peter Neigel, Raj Bhalwankar, Daniel Kondermann, Sebastian J. Vollmer arxiv

Vision language models (VLMs) excel at zero-shot visual classification, but their performance on fine-grained tasks and large hierarchical label spaces is understudied. This paper investigates whether structured, tree-based reasoning can enhance VLM performance. We introduce a framework that decomposes classification into interpretable decisions using decision trees and evaluates it on fine-grained (GTSRB) and coarse-grained (CIFAR-10) datasets. Although the model achieves 98.2% accuracy in understanding the tree knowledge, tree-based reasoning consistently underperforms standard zero-shot prompting. We also explore enhancing the tree prompts with LLM-generated classes and image descriptions to improve alignment. The added description enhances the performance of the tree-based and zero-shot methods. Our findings highlight limitations of structured reasoning in visual classification and offer insights for designing more interpretable VLM systems.

📄 PDF Abstract BibTeX arXiv:2509.09732

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Reasoning over Hierarchical Question Decomposition Tree for Explainable Question Answering

2023-05-24 · Jiajie Zhang, Shulin Cao, Tingjia Zhang, Xin Lv 외

Explainable question answering (XQA) aims to answer a given question and provide an explanation why the answer is selected. Existing XQA methods focus on reasoning on a single knowledge source, e.g., structured knowledge…

Question Answering

An Enhanced Projection Pursuit Tree Classifier with Visual Methods for Assessing Algorithmic Improvements

2026-02-24 · Natalia da Silva, Dianne Cook, Eun-Kyung Lee arxiv

This paper presents enhancements to the projection pursuit tree classifier and visual diagnostic methods for assessing their impact in high dimensions. The original algorithm uses linear combinations of variables in a tr…

AI vs Human Expert Reasoning: Assessing Agreements in Building Typology Predictions based on Street View Imagery

2026-07-16 · Zahratu Shabrina, Muhammad Asa, Jin Rui, Lu Yin 외 arxiv

This research investigates the potential of Vision-Language Models (VLMs) to infer building typologies: Construction, Current Use, and Storeys from Google Street View (GSV) images. Predictions generated by VLMs are compa…

MARIC: Multi-Agent Reasoning for Image Classification

2025-09-18 · Wonduk Seo, Minhyeong Yu, Hyunjin An, Seunghyun Lee arxiv

Image classification has traditionally relied on parameter-intensive model training, requiring large-scale annotated datasets and extensive fine tuning to achieve competitive performance. While recent vision language mod…

Image ClassificationVisual Reasoning

Automatically Inferring Teachers' Geometric Content Knowledge: A Skills Based Approach

2026-04-15 · Ziv Fenigstein, Kobi Gal, Avi Segal, Osama Swidan 외 arxiv

Assessing teachers' geometric content knowledge is essential for geometry instructional quality and student learning, but difficult to scale. The Van Hiele model characterizes geometric reasoning through five hierarchica…

Multi-Task Learning