5 papers
Does VLA Even Know the Basics? Measuring Commonsense and World Knowledge Retention in Vision-Language-Action Models
Nikita Kachaev, Andrey Moskalenko, Matvey Skripkin +10
Embodied Vision-Language-Action (VLA) models are typically obtained by fine-tuning powerful pretrained VLMs on robotics data, yet it is unclear how much commonsense and factual kno…
BREPS: Bounding-Box Robustness Evaluation of Promptable Segmentation
Andrey Moskalenko, Danil Kuznetsov, Irina Dudko +6
Promptable segmentation models such as SAM have established a powerful paradigm, enabling strong generalization to unseen objects and domains with minimal user input, including poi…
SPARTA: Evaluating Reasoning Segmentation Robustness through Black-Box Adversarial Paraphrasing in Text Autoencoder Latent Space
Viktoriia Zinkovich, Anton Antonov, Andrei Spiridonov +6
Multimodal large language models (MLLMs) have shown impressive capabilities in vision-language tasks such as reasoning segmentation, where models generate segmentation masks based…
Bring the Apple, Not the Sofa: Impact of Irrelevant Context in Embodied AI Commands on VLA Models
Daria Pugacheva, Andrey Moskalenko, Denis Shepelev +3
Vision Language Action (VLA) models are widely used in Embodied AI, enabling robots to interpret and execute language instructions. However, their robustness to natural language va…
RClicks: Realistic Click Simulation for Benchmarking Interactive Segmentation
Anton Antonov, Andrey Moskalenko, Denis Shepelev +4
The emergence of Segment Anything (SAM) sparked research interest in the field of interactive segmentation, especially in the context of image editing tasks and speeding up data an…