Less Detail, Better Answers: Degradation-Driven Prompting for VQA
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 speci
Weijie Wang*Zeyu ZhangYefei He Bohan ZhuangState Key Lab of CAD&CG, Zhejiang University 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.…
saved by
related reading
- GitHub - FusionBrainLab/Vision_GRPOgithub.com
- Seeing Is Not Reasoning: How VLMs and Their Benchmarks Lean on Textharvey-fin.github.io
- State of Vision-Language-Action (VLA) Research at ICLR 2026 – Moritz Reussmbreuss.github.io
- 2403.09611.pdfarxiv.org
- TraveLER: A Modular Multi-LMM Agent Framework for Video Question-Answeringarxiv.org
- Explore | alphaXivalphaxiv.org
- HALVA: Hallucination Attenuated Language and Vision Assistantresearch.google
- DeepSeek-R1arxiv.org
- Circuit Tracing in Vision–Language Models:Understanding the Internal Mechanisms of Multimodal Thinkingarxiv.org
- Video models are zero-shot learners and reasonersarxiv.org
- [2510.13626] LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Modelsarxiv.org
- MMHal Bencharxiv.org