Benchmark RadarAI BENCHMARK PROFILE
VQABench
Evaluates 12 image preprocessing techniques for cloud VLM-based visual question answering across 3 VQA datasets and 4 commercial models, measuring accuracy, cost, and latency.
- Released
- 2026-08-08
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Assesses the impact of client-side preprocessing on cost-quality trade-offs for offloaded VQA inference, offering practical guidance for system design.
Motivation
Vision-language models (VLMs) are becoming a practical backend for mobile visual question answering (VQA) systems, enabling smartphones and smart glasses to answer users' questions about the physical world.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.