Skip to yearly menu bar Skip to main content


Exploring Question Decomposition for Zero-Shot VQA

Zaid Khan · Vijay Kumar B G · Samuel Schulter · Manmohan Chandraker · Yun Fu

Great Hall & Hall B1+B2 (level 1) #124
[ ]
Wed 13 Dec 8:45 a.m. PST — 10:45 a.m. PST


Visual question answering (VQA) has traditionally been treated as a single-step task where each question receives the same amount of effort, unlike natural human question-answering strategies. We explore a question decomposition strategy for VQA to overcome this limitation. We probe the ability of recently developed large vision-language models to use human-written decompositions and produce their own decompositions of visual questions, finding they are capable of learning both tasks from demonstrations alone.However, we show that naive application of model-written decompositions can hurt performance.We introduce a model-driven selective decomposition approach for second-guessing predictions and correcting errors, and validate its effectiveness on eight VQA tasks across three domains, showing consistent improvements in accuracy, including improvements of >20% on medical VQA datasets and boosting the zero-shot performance of BLIP-2 above chance on a VQA reformulation of the challenging Winoground task. Project Site:

Chat is not available.