Publications

Abstract

While research in natural language processing has progressed significantly in creative language generation, the question of whether language models can interpret the intended meaning of creative language largely remains unanswered. Poetry as a creative art form has existed for generations, and summarization of such content requires deciphering the figurative patterns to find out the actual intent and message of the poet. This task can provide the researchers an opportunity to evaluate the creative language interpretation capacity of the language models. Unlike typical text, summarization of poems is a challenging task as poems carry a deeper meaning, which can be easily lost if only the literal meaning is considered. That being said, we propose a new task in the field of natural language understanding called 'Poem Summarization'. As a starting, we propose the first-ever dataset for this task, named 'PoemSum', consisting of 3011 samples of poetry and its corresponding summarized interpretation in the English language. We have benchmarked the performance of different state-of-the-art summarization models and provided observations on their limitations. The dataset and all relevant code used in this work have been made publicly available.

Abstract

Ensuring global food sufficiency and security is one of the prime challenges of the twenty-first century. The most effective approach to tackle this challenge is to ensure a healthy agricultural ecosystem. A potential barrier, in this case, would be different diseases that commonly infest and cause great damage to the production. To keep plants disease-free, most countries still rely on human intervention-based approaches. One issue with the mentioned approach is that farmers don't get the help they need at the right time owing to manpower shortages. This paves the way for the development and implementation of automated mechanisms to detect and classify plant disease. Using heavy-weight convolutional neural network or CNN-driven solutions is often not practical as farmers are not equipped with devices capable of running such heavy applications. This is why lightweight CNN architectures capable of operating mobile and embedded devices are crucial. In this work, we present a comparative analysis and overview of different efficient CNN-based methodologies proposed for plant disease classification. Moreover, we fine-tuned off-the-shelf state-of-the-art efficient CNN architectures using transfer learning to analyze and determine the right balance of model size and accuracy.