6 papers
GUI-Perturbed: Domain Randomization Reveals Systematic Brittleness in GUI Grounding Models
Yangyue Wang, Harshvardhan Sikka, Yash Mathur +3
GUI grounding models report over 85% accuracy on standard benchmarks, yet drop 27-56 percentage points when instructions require spatial reasoning rather than direct element naming…
Benchmarking the Generality of Vision-Language-Action Models
Pranav Guruprasad, Sudipta Chowdhury, Harsh Sikka +6
Generalist multimodal agents are expected to unify perception, language, and control - operating robustly across diverse real world domains. However, current evaluation practices r…
An Open-Source Software Toolkit & Benchmark Suite for the Evaluation and Adaptation of Multimodal Action Models
Pranav Guruprasad, Yangyue Wang, Sudipta Chowdhury +2
Recent innovations in multimodal action models represent a promising direction for developing general-purpose agentic systems, combining visual understanding, language comprehensio…
Benchmarking Vision, Language, & Action Models in Procedurally Generated, Open Ended Action Environments
Pranav Guruprasad, Yangyue Wang, Sudipta Chowdhury +2
Vision-language-action (VLA) models represent an important step toward general-purpose robotic systems by integrating visual perception, language understanding, and action executio…
Benchmarking Vision, Language, & Action Models on Robotic Learning Tasks
Pranav Guruprasad, Harshvardhan Sikka, Jaewoo Song +2
Vision-language-action (VLA) models represent a promising direction for developing general-purpose robotic systems, demonstrating the ability to combine visual understanding, langu…
KULCQ: An Unsupervised Keyword-based Utterance Level Clustering Quality Metric
Pranav Guruprasad, Negar Mokhberian, Nikhil Varghese +2
Intent discovery is crucial for both building new conversational agents and improving existing ones. While several approaches have been proposed for intent discovery, most rely on…