Search

Your search keyword '"Vyas Kaustubh"' showing total 18 results
Start Over Author "Vyas Kaustubh"
18 results on '"Vyas Kaustubh"'

1. Phase-matched third-harmonic generation in silicon nitride waveguides

Author

Vijayakumar Surendar, Vyas Kaustubh, Espinosa Daniel H. G., Reshef Orad, Song Meiting, Awan Kashif Masud, Choudhary Saumya, Cardenas Jaime, Boyd Robert W., and Dolgaleva Ksenia

Subjects
harmonic generation, waveguide, frequency conversion, silicon nitride, nonlinear optics, Physics, QC1-999
Abstract

Third-harmonic generation (THG) in silicon nitride waveguides is an ideal source of coherent visible light, suited for ultrafast pulse characterization, telecom signal monitoring and self-referenced comb generation due to its relatively large nonlinear susceptibility and CMOS compatibility. We demonstrate third-harmonic generation in silicon nitride waveguides where a fundamental transverse mode at 1,596 nm is phase-matched to a TM02 mode at 532 nm, confirmed by the far-field image. We experimentally measure the waveguide width-dependent phase-matched wavelength with a peak-power-normalized conversion efficiency of 5.78 × 10−7 %/W2 over a 660-μm-long interaction length.

Published
2024
Full Text
View/download PDF

2. Improving Retrieval-augmented Text-to-SQL with AST-based Ranking and Schema Pruning

Author

Shen, Zhili, Vougiouklis, Pavlos, Diao, Chenxin, Vyas, Kaustubh, Ji, Yuanyi, and Pan, Jeff Z.

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence, Computer Science - Databases
Abstract

We focus on Text-to-SQL semantic parsing from the perspective of retrieval-augmented generation. Motivated by challenges related to the size of commercial database schemata and the deployability of business intelligence solutions, we propose $\text{ASTReS}$ that dynamically retrieves input database information and uses abstract syntax trees to select few-shot examples for in-context learning. Furthermore, we investigate the extent to which an in-parallel semantic parser can be leveraged for generating approximated versions of the expected SQL queries, to support our retrieval. We take this approach to the extreme--we adapt a model consisting of less than $500$M parameters, to act as an extremely efficient approximator, enhancing it with the ability to process schemata in a parallelised manner. We apply $\text{ASTReS}$ to monolingual and cross-lingual benchmarks for semantic parsing, showing improvements over state-of-the-art baselines. Comprehensive experiments highlight the contribution of modules involved in this retrieval-augmented generation setting, revealing interesting directions for future work., Comment: EMNLP 2024 Main

Published
2024

Catalog

Books, media, physical & digital resources
See catalog results