10 papers
Can These Views Be One Scene? Evaluating Multiview 3D Consistency when 3D Foundation Models Hallucinate
Soumava Paul, Prakhar Kaushik, Alan Yuille
Multiview 3D evaluation assumes that the images being scored are observations of one static 3D scene. This assumption can fail in NVS and sparse-view reconstruction: inputs or gene…
Shared LoRA Subspaces for almost Strict Continual Learning
Prakhar Kaushik, Ankit Vaidya, Shravan Chaudhari +2
Adapting large pretrained models to new tasks efficiently and continually is crucial for real-world deployment but remains challenging due to catastrophic forgetting and the high c…
EigenLoRAx: Recycling Adapters to Find Principal Subspaces for Resource-Efficient Adaptation and Inference
Prakhar Kaushik, Ankit Vaidya, Shravan Chaudhari +1
The rapid growth of large models has raised concerns about their environmental impact and equity in accessibility due to significant computational costs. Low-Rank Adapters (LoRA) o…
Name That Part: 3D Part Segmentation and Naming
Soumava Paul, Prakhar Kaushik, Ankit Vaidya +2
We address semantic 3D part segmentation: decomposing objects into parts with meaningful names. While datasets exist with part annotations, their definitions are inconsistent acros…
The Universal Weight Subspace Hypothesis
Prakhar Kaushik, Shravan Chaudhari, Ankit Vaidya +2
We show that deep neural networks trained across diverse tasks exhibit remarkably similar low-dimensional parametric subspaces. We provide the first large-scale empirical evidence…
Perceptual Taxonomy: Evaluating and Guiding Hierarchical Scene Reasoning in Vision-Language Models
Jonathan Lee, Xingrui Wang, Jiawei Peng +9
We propose Perceptual Taxonomy, a structured process of scene understanding that first recognizes objects and their spatial configurations, then infers task-relevant properties suc…