HCNQA: Enhancing 3D VQA with Hierarchical Concentration Narrowing Supervision

TL;DR

HCNQA introduces a hierarchical supervision approach for 3D VQA that guides models through phased concentration narrowing, improving reasoning pathways and overall performance.

Contribution

The paper proposes a novel hierarchical concentration narrowing supervision method for 3D VQA, enhancing reasoning pathway development and model accuracy.

Findings

Improved reasoning pathways in 3D VQA models.

Enhanced accuracy over existing answer-centric methods.

Effective supervision at key reasoning checkpoints.

Abstract

3D Visual Question-Answering (3D VQA) is pivotal for models to perceive the physical world and perform spatial reasoning. Answer-centric supervision is a commonly used training method for 3D VQA models. Many models that utilize this strategy have achieved promising results in 3D VQA tasks. However, the answer-centric approach only supervises the final output of models and allows models to develop reasoning pathways freely. The absence of supervision on the reasoning pathway enables the potential for developing superficial shortcuts through common patterns in question-answer pairs. Moreover, although slow-thinking methods advance large language models, they suffer from underthinking. To address these issues, we propose \textbf{HCNQA}, a 3D VQA model leveraging a hierarchical concentration narrowing supervision method. By mimicking the human process of gradually focusing from a broad area to specific objects while searching for answers, our method guides the model to perform three phases of concentration narrowing through hierarchical supervision. By supervising key checkpoints on a general reasoning pathway, our method can ensure the development of a rational and effective reasoning pathway. Extensive experimental results demonstrate that our method can effectively ensure that the model develops a rational reasoning pathway and performs better. The code is available at https://github.com/JianuoZhu/HCNQA.