IC-018Object information is localized to specific visual tokens in LLaVA-1.5

Clement Neo, Luke Ong, Philip Torr, Mor Geva, David Krueger, Fazl Barez

SourceTowards Interpreting Visual Information Processing in Vision-Language Models

The study investigates whether object-specific information in the visual tokens of the language model component of LLaVA-1.5 is localized or dispersed. Ablation experiments were conducted where tokens corresponding to the object's spatial location were removed and replaced with a mean embedding. This led to a significant drop in the model's ability to identify the object across three tasks (generative description, binary polling, and VQA). The drop was substantially larger than when ablating an equivalent number of random tokens or tokens identified as important via Integrated Gradients, supporting the claim of localization.

Evidence
interventional
Key metric
For generative description: 71.79% decrease (for +1 buffer, 33.4 tokens); For polling: 51.28% decrease; For VQA: 86.67% decrease.
Caveat
The results are for object identification tasks and on the LLaVA family of models. The study only investigates a subset of 100 images for the VQA task, and the filtering process for the dataset may introduce bias.
Model
LLaVA-1.5 / LLaVA-v1.5, LLaVA-Phi
Datasets
MS COCO / COCO / COCO 2014 / COCO 2017 / COCO 20k / COCO-it / COCO-wl [eval], ImageNet-1k / ImageNet / ImageNet-1k-val / ImageNet-Val [eval]
Methods
Integrated Gradients / Integral of gradients [compared-to]
Extraction
automatic-extraction