Adding answer-preserving visual or relational modifiers to visual questions lowers accuracy of GPT-4o, Gemini-1.5-Flash, and Claude-3.5-Sonnet on VQA v2.0.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Analyzing the Sensitivity of Vision Language Models in Visual Question Answering
Adding answer-preserving visual or relational modifiers to visual questions lowers accuracy of GPT-4o, Gemini-1.5-Flash, and Claude-3.5-Sonnet on VQA v2.0.