6 papers
Discrete Langevin-Inspired Posterior Sampling
Chaitanya Amballa, Sattwik Basu, Jorge Vančo Sampedro +1
We study posterior sampling for inverse problems in discrete state spaces using discrete diffusion models as generative priors. While continuous diffusion models have become widely…
AVMeme Exam: A Multimodal Multilingual Multicultural Benchmark for LLMs' Contextual and Cultural Knowledge and Thinking
Xilin Jiang, Qiaolin Wang, Junkai Wu +30
Internet audio-visual clips convey meaning through time-varying sound and motion, which extend beyond what text alone can represent. To examine whether AI models can understand suc…
Learning Energy-based Variational Latent Prior for VAEs
Debottam Dutta, Chaitanya Amballa, Zhongweiyang Xu +2
Variational Auto-Encoders (VAEs) are known to generate blurry and inconsistent samples. One reason for this is the "prior hole" problem. A prior hole refers to regions that have hi…
Contrastive Diffusion Guidance for Spatial Inverse Problems
Sattwik Basu, Chaitanya Amballa, Zhongweiyang Xu +3
We consider a class of inverse problems characterized by forward operators that are partially specified, non-smooth, and non-differentiable. Although generative inverse solvers hav…
Explicit Context-Driven Neural Acoustic Modeling for High-Fidelity RIR Generation
Chen Si, Qianyi Wu, Chaitanya Amballa +1
Realistic sound simulation plays a critical role in many applications. A key element in sound simulation is the room impulse response (RIR), which characterizes how sound propagate…
Can NeRFs See without Cameras?
Chaitanya Amballa, Sattwik Basu, Yu-Lin Wei +3
Neural Radiance Fields (NeRFs) have been remarkably successful at synthesizing novel views of 3D scenes by optimizing a volumetric scene function. This scene function models how op…