D. Paudel

This work presents EarthMind, a novel vision-language framework for multi-granular and multi-sensor EO data understanding and outperforms existing methods on multiple public EO benchmarks, showcasing its potential to handle both multi-granular and multi-sensor challenges in a unified framework.

Diffusion models can lose diversity when only later timesteps are updated during reward optimization, a finding that challenges prior assumptions and motivates a new training paradigm.

CoRef-GS is proposed, a cooperative referring Gaussian splatting framework that constructs local open-vocabulary instance-aware Gaussian maps, then aligns partially overlapping maps with a cross-agent alignment module by geometric and semantic consistency, and grounds queries using a view-conditioned mask relation graph.

X-MULTI not only synthesizes unobserved imaging factor combinations but also reveals that existing evaluation metrics can misrepresent disentanglement quality.

Achieving sub-pixel accuracy with the lowest latency, ODF Motion Estimation transforms event-based motion tasks while enabling real-time applications like image deblurring and efficient tracking.

iARCS transforms 3D scene generation by ensuring that synthetic environments meet essential functional constraints while maintaining diversity and realism.

Over 1,500 submissions revealed stark differences in model performance across diverse domains, highlighting the challenges of generalizing egocentric video understanding.

Image-size agnostic vision transformers are now a practical reality, thanks to a new self-supervised pretraining method that maintains constant computational cost regardless of input resolution.

Autonomous vehicles can now better identify the unexpected, thanks to a new method that boosts out-of-distribution detection by up to 20% without retraining.

Forget expensive teacher models and manual labeling: a base VLM paired with OpenStreetMap data can annotate itself for remote sensing tasks, achieving state-of-the-art performance at a fraction of the cost.

EarthMind demonstrates that hierarchical cross-modal attention across optical and SAR data significantly boosts MLLM performance on Earth Observation tasks, outperforming models limited to single-sensor inputs.