Searcharxiv⌕ Search

arXiv subjects

Rishant Sharma

Publications and source records attributed to Rishant Sharma.

2 recordsLinked to original sources

Out-of-time-order Correlators in Volcano potentials

The out-of-time-order correlator (OTOC) has emerged as an important diagnostic for quantum information scrambling and a key signature of quantum chaos. In this work, we discuss out-of-time-order correlators, analytically and numerically, for the so-called Volcano potentials. This family of potentials parametrically interpolates between bounded wells and confining walls, and includes Pöschl-Teller as a specific case. Through a combination of numerical diagonalization and semiclassical analysis, we investigate various regimes of the potential parameter regimes containing regions of both positive and negative curvatures. Moreover, we demonstrate in this case that short-time exponential growth of the OTOC does not necessarily signal global quantum chaos; instead, it correlates with the sampling of negative curvature regions by excited states.

quant-ph↗

Lights, Camera, Consistency: A Multistage Pipeline for Character-Stable AI Video Stories

Generating long, cohesive video stories with consistent characters is a significant challenge for current text-to-video AI. We introduce a method that approaches video generation in a filmmaker-like manner. Instead of creating a video in one step, our proposed pipeline first uses a large language model to generate a detailed production script. This script guides a text-to-image model in creating consistent visuals for each character, which then serve as anchors for a video generation model to synthesize each scene individually. Our baseline comparisons validate the necessity of this multi-stage decomposition; specifically, we observe that removing the visual anchoring mechanism results in a catastrophic drop in character consistency scores (from 7.99 to 0.55), confirming that visual priors are essential for identity preservation. Furthermore, we analyze cultural disparities in current models, revealing distinct biases in subject consistency and dynamic degree between Indian vs Western-themed generations.

cs.CV↗