GroundCount: Grounding Vision-Language Models with Object Detection for Mitigating Counting Hallucinations

TL;DR

GroundCount enhances vision-language models' counting accuracy by integrating object detection for explicit spatial grounding, significantly reducing hallucinations and inference time across multiple architectures.

Contribution

It introduces GroundCount, a novel framework that combines object detection with VLMs to mitigate counting hallucinations through explicit grounding and prompt-based augmentation.

Findings

Achieves up to 81.3% counting accuracy, a 6.6pp improvement.

Reduces inference time by 22% by eliminating hallucination-driven reasoning.

Explicit grounding outperforms implicit feature fusion methods.

Abstract

Vision Language Models (VLMs) exhibit persistent hallucinations in counting tasks, with accuracy substantially lower than other visual reasoning tasks (excluding sentiment). This phenomenon persists even in state-of-the-art reasoning-capable VLMs. Conversely, CNN-based object detection models (ODMs) such as YOLO excel at spatial localization and instance counting with minimal computational overhead. We propose GroundCount, a framework that augments VLMs with explicit spatial grounding from ODMs to mitigate counting hallucinations. In the best case, our prompt-based augmentation strategy achieves 81.3% counting accuracy on the best-performing model (Ovis2.5-2B) - a 6.6pp improvement - while reducing inference time by 22% through elimination of hallucination-driven reasoning loops for stronger models. We conduct comprehensive ablation studies demonstrating that positional encoding is a critical component, being beneficial for stronger models but detrimental for weaker ones. Confidence scores, by contrast, introduce noise for most architectures and their removal improves performance in four of five evaluated models. We further evaluate feature-level fusion architectures, finding that explicit symbolic grounding via structured prompts outperforms implicit feature fusion despite sophisticated cross-attention mechanisms. Our approach yields consistent improvements across four of five evaluated VLM architectures (6.2--7.5pp), with one architecture exhibiting degraded performance due to incompatibility between its iterative reflection mechanisms and structured prompts. These results suggest that counting failures stem from fundamental spatial-semantic integration limitations rather than architecture-specific deficiencies, while highlighting the importance of architectural compatibility in augmentation strategies.