Numbers are essential components of text, like any other word tokens, from which natural language processing (NLP) models are built and deployed. Though numbers are typically not accounted for distinctly in most NLP tasks, there is still an underlying amount of numeracy already exhibited by NLP models. For instance, in named entity recognition (NER), numbers are not treated as an entity with distinct tags. In this work, we attempt to tap the potential of state-of-the-art language models and transfer their ability to boost performance in related downstream tasks dealing with numbers. Our proposed classification of numbers into entities helps NLP models perform well on several tasks, including a handcrafted Fill-In-The-Blank (FITB) task and on question answering, using joint embeddings, outperforming the BERT and RoBERTa baseline classification.

Vivek Subramanian, Matthew Engelhard, Sam Berchuck, Liqun Chen, Ricardo Henao, Lawrence Carin. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

Dhanasekar Sundararaman, Shijing Si, Vivek Subramanian, Guoyin Wang, Devamanyu Hazarika, Lawrence Carin. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2020.

The realization of a high-speed printing technique with high resolution and pattern fidelity is critical to making printed electronics a viable technology for electronics manufacturing. The printing requirements of printed electronics are substantially different that those of graphic arts. To make printed electronics a reality, it is necessary to deliver high resolution, good reproducibility, excellent pattern fidelity, high process throughput, and compatibility with the requisite semiconductor, dielectric, and conductor inks. In this paper, we review the physics of pattern formation from pixelated primitives, such as those that exist during inkjet and gravure printing, and will show how control of drop merging and drying can be used to produce high-fidelity shapes, including lines, squares, and intersections. We additionally discuss the physical underpinnings of gravure printing and inkjet printing, and show how these techniques can be scaled to produce high-fidelity highly scaled patterns, including sub-2 micron features at printing speeds of ~1 m/s. Finally, in conjunction with high-performance materials, we describe our realization of high-performance fully printed transistors on plastic, offering high-switching speed, excellent process throughput, and good fidelity over large areas.

In their first electronics course, many students find operational amplifiers, analog filters and sensor interface circuitry perplexing and daunting. The purpose of this paper is to present a circuit that addresses these pitfalls. A simplified electroencephelogram (EEG) circuit that is interfaced to a digital backend is proposed. The completed circuit involves using instrumentation amplifiers and filters for the EEG interface. The digital backend helps analyze EEG data on the computer.

Printed electronics provides a promising potential pathway toward the realization of ultralow-cost RFID tags for item-level tracking of consumer goods. Here, we report on our progress in developing materials, processes, and devices for the realization of ultralow-cost printed RFID tags. Using printed nanoparticle patterns that are subsequently sintered at plastic-compatible temperatures, low-resistance interconnects and passive components have been realized. Simultaneously, printed transistors with mobilities >10/sup -1/ cm/sup 2//V-s have been realized using novel pentacene and oligothiophene precursors for pMOS and ZnO nanoparticles for nMOS. AC performance of these devices is adequate for 135-kHz RFID, though significant work remains to be done to achieve 13.56-MHz operation.