Stories are diverse and highly personalized, resulting in a large possible output space for story generation. Existing end-to-end approaches produce monotonous stories because they are limited to the vocabulary and knowledge in a single training dataset. This paper introduces KG-Story, a three-stage framework that allows the story generation model to take advantage of external Knowledge Graphs to produce interesting stories. KG-Story distills a set of representative words from the input prompts, enriches the word set by using external knowledge graphs, and finally generates stories based on the enriched word set. This distill-enrich-generate framework allows the use of external resources not only for the enrichment phase, but also for the distillation and generation phases. In this paper, we show the superiority of KG-Story for visual storytelling, where the input prompt is a sequence of five photos and the output is a short story. Per the human ranking evaluation, stories generated by KG-Story are on average ranked better than that of the state-of-the-art systems. Our code and output stories are available at https://github.com/zychen423/KE-VIST.
A fully integrated, miniaturized, low-power frequency-modulated continuous wave (FMCW) multifunction chip realized by typical 1P6M 0.18 gm deep n-well CMOS technology is presented for the first time. The multifunction chip consists of VCO, buffer amplifier, 3-dB power divider, isolators, driving amplifiers, mixer, low-noise amplifier, attenuator, etc., necessary for carrying out the X-band RF signal processing of the FMCW signals interfaced to dual antenna arrays. The chip real estate measures 2.4 mm by 1.3 mm. The entire FMCW chip design is based on the synthetic complementary-conductingstrips (CCS) quasi-TEM transmission line. The transmitter output is 3.5 dBm for frequencies between 9.5-11.0 GHz and maximum tuning bandwidth is nearly 150 MHz. The receiver channel has conversion gain of 6 dB. The calculated range is in good agreement with the measurement data.
Stories are diverse and highly personalized, resulting in a large possible output space for story generation. Existing endto-end approaches produce monotonous stories because they are limited to the vocabulary and knowledge in a single training dataset. This paper introduces KG-Story, a three-stage framework that allows the story generation model to take advantage of external Knowledge Graphs to produce interesting stories. KG-Story distills a set of representative words from the input prompts, enriches the word set by using external knowledge graphs, and finally generates stories based on the enriched word set. This distill-enrich-generate framework allows the use of external resources not only for the enrichment phase, but also for the distillation and generation phases. In this paper, we show the superiority of KG-Story for visual storytelling, where the input prompt is a sequence of five photos and the output is a short story. Per the human ranking evaluation, stories generated by KG-Story are on average ranked better than that of the state-of-theart systems. Our code and output stories are available at https://github.com/zychen423/KE-VIST.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.