6 papers
Multimodal Language Models Cannot Spot Spatial Inconsistencies
Om Khangaonkar, Hadi J. Rad, Hamed Pirsiavash
Spatial consistency is a fundamental property of the visual world and a key requirement for models that aim to understand physical reality. Despite recent advances, multimodal larg…
Vector-Quantized Soft Label Compression for Dataset Distillation
Ali Abbasi, Ashkan Shahbazi, Hamed Pirsiavash +1
Dataset distillation is an emerging technique for reducing the computational and storage costs of training machine learning models by synthesizing a small, informative subset of da…
LaVR: Scene Latent Conditioned Generative Video Trajectory Re-Rendering using Large 4D Reconstruction Models
Mingyang Xie, Numair Khan, Tianfu Wang +8
Given a monocular video, the goal of video re-rendering is to generate views of the scene from a novel camera trajectory. Existing methods face two distinct challenges. Geometrical…
The Path Not Taken: RLVR Provably Learns Off the Principals
Hanqing Zhu, Zhenyu Zhang, Hanxian Huang +11
Reinforcement Learning with Verifiable Rewards (RLVR) reliably improves the reasoning performance of large language models, yet it appears to modify only a small fraction of parame…
gen2seg: Generative Models Enable Generalizable Instance Segmentation
Om Khangaonkar, Hamed Pirsiavash
By pretraining to synthesize coherent images from perturbed inputs, generative models inherently learn to understand object boundaries and scene compositions. How can we repurpose…
MoIN: Mixture of Introvert Experts to Upcycle an LLM
Ajinkya Tejankar, KL Navaneet, Ujjawal Panchal +2
The goal of this paper is to improve (upcycle) an existing large language model without the prohibitive requirements of continued pre-training of the full-model. The idea is to spl…