10 papers
The Planning Limits of Latent World Models
Ali Alrasheed, Basim Azam, Naveed Akhtar
World models offer a promising way to help robots understand how the physical world evolves and plan complex behaviours through imagination. Yet existing studies mainly demonstrate…
MEND: Label-Free Detection, Localisation, and Correction of Latent Hallucination in World Models
Ali J Alrasheed, Aryan Yazdan Parast, Basim Azam +2
World Models are appearing as the next major frontier in computer vision. However, their robustness is currently largely unexplored. We identify the phenomenon of hallucination in…
InstructMixup: Instruction-Guided Salient Patch Editing for Robust Data Augmentation
Khawar Islam, Arif Mahmood, Xin Jin +1
In image and video technologies, data augmentation is widely used to improve the generalization of deep visual models, and mixup-based strategies that interpolate between samples h…
Reasoning as a Double-Edged Sword: Architecture and Cross-Stage Robustness in Vision-Language-Action Models
Tuan Duong Trinh, Naveed Akhtar, Basim Azam
Does adding a reasoning step make a Vision-Language-Action (VLA) model more robust to perturbation? Intuitively, a policy that reasons before acting should absorb a perturbed input…
-FracMix: Label-Preserving Self-Saliency Mixup Augmentation
Khawar Islam, Arif Mahmood, Xin Jin +1
Data augmentation is known to improve generalization of deep visual models. Recent methods favor mixup strategies that generate interpolated samples to improve model performance. H…
Invoice Haystack: Benchmarking Document Retrieval and Visual Question Answering Under Strong Visual Homogeneity
Heethanjan Kanagalingam, Thenukan Pathmanathan, Mokeeshan Vathanakumar +3
Vision Language Models have achieved near-human performance on single-document Visual Question Answering, yet their effectiveness degrades significantly when retrieving information…
Symmetry-Aware 9D Pose Estimation with Sim(3)-Consistent Feature and Spherical Inception Convolution
Panfei Cheng, Hongshan Yu, Wenrui Chen +3
Object pose estimation is a fundamental problem for an agent system to perceive or manipulate objects in images or videos. However, current instance-level methods struggle with gen…
Latent Video Prediction for World Modeling: An Evaluation Uncovering Intriguing Favorable Evidence
Ali J Alrasheed, Aryan Yazdan Parast, Basim Azam +2
Self-supervised video models are increasingly framed as world models, yet they are still evaluated almost entirely on clean video and reported as a final task score, obscuring how…
Mitigating Memorization in Text-to-Image Diffusion via Region-Aware Prompt Augmentation and Multimodal Copy Detection
Yunzhuo Chen, Jordan Vice, Naveed Akhtar +2
State-of-the-art text-to-image diffusion models can produce impressive visuals but may memorize and reproduce training images, creating copyright and privacy risks. Existing prompt…
Altered Thoughts, Altered Actions: Reasoning Chain as Control Surface for a Vision-Language-Action Policy
Tuan Duong Trinh, Naveed Akhtar, Basim Azam +2
Vision-language-action policies map camera images and natural-language instructions to a robot's motor actions. Some of these policies are designed to reason in text before acting,…