Paper Detail
Haoxuan Han, Weijie Wang, Zeyu Zhang, Yefei He, Bohan Zhuang
Recent advancements in Vision-Language Models (VLMs) have significantly pushed the boundaries of Visual Question Answering (VQA).However,high-resolution details can sometimes become noise that leads to hallucinations or reasoning errors. In this paper,we propose Degradation-Driven Prompting (DDP), a novel framework that improves VQA performance by strategically reducing image fidelity to force models to focus on essential structural information. We evaluate DDP across two distinct tasks. Physical attributes targets images prone to human misjudgment, where DDP employs a combination of 80p downsampling, structural visual aids (white background masks and orthometric lines), and In-Context Learning (ICL) to calibrate the model's focus. Perceptual phenomena addresses various machine-susceptible visual anomalies and illusions, including Visual Anomaly (VA), Color (CI), Motion(MI),Gestalt (GI), Geometric (GSI), and Visual Illusions (VI).For this task, DDP integrates a task-classification stage with specialized tools such as blur masks and contrast enhancement alongside downsampling. Our experimental results demonstrate that less is more: by intentionally degrading visual inputs and providing targeted structural prompts, DDP enables VLMs to bypass distracting textures and achieve superior reasoning accuracy on challenging visual benchmarks.
No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.
No ranking explanation is available yet.
No tags.
@misc{han2026less,
title = {Less Detail, Better Answers: Degradation-Driven Prompting for VQA},
author = {Haoxuan Han and Weijie Wang and Zeyu Zhang and Yefei He and Bohan Zhuang},
year = {2026},
abstract = {Recent advancements in Vision-Language Models (VLMs) have significantly pushed the boundaries of Visual Question Answering (VQA).However,high-resolution details can sometimes become noise that leads to hallucinations or reasoning errors. In this paper,we propose Degradation-Driven Prompting (DDP), a novel framework that improves VQA performance by strategically reducing image fidelity to force models to focus on essential structural information. We evaluate DDP across two distinct tasks. Physica},
url = {https://huggingface.co/papers/2604.04838},
keywords = {Vision-Language Models, Visual Question Answering, image fidelity, structural information, In-Context Learning, visual anomalies, visual illusions, downsampling, structural visual aids, blur masks, contrast enhancement, task-classification stage, code available, huggingface daily},
eprint = {2604.04838},
archiveprefix = {arXiv},
}
{}