The configurational information in sentences of a free word order language such as Sanskrit is of limited use. Thus, the context of the entire sentence will be desirable even for basic processing tasks such as word segmentation. We propose a structured prediction framework that jointly solves the word segmentation and morphological tagging tasks in Sanskrit. We build an energy based model where we adopt approaches generally employed in graph based parsing techniques (McDonald et al., 2005a;Carreras, 2007). Our model outperforms the state of the art with an F-Score of 96.92 (percentage improvement of 7.06%) while using less than one tenth of the task-specific training data. We find that the use of a graph based approach instead of a traditional lattice-based sequential labelling approach leads to a percentage gain of 12.6% in F-Score for the segmentation task. 1

Data augmentation is a widely employed technique to alleviate the problem of data scarcity. In this work, we propose a prompting-based approach to generate labelled training data for intent classification with off-the-shelf language models (LMs) such as GPT-3. An advantage of this method is that no task-specific LM-fine-tuning for data generation is required; hence the method requires no hyper-parameter tuning and is applicable even when the available training data is very scarce. We evaluate the proposed method in a few-shot setting on four diverse intent classification tasks. We find that GPT-generated data significantly boosts the performance of intent classifiers when intents in consideration are sufficiently distinct from each other. In tasks with semantically close intents, we observe that the generated data is less helpful. Our analysis shows that this is because GPT often generates utterances that belong to a closely-related intent instead of the desired one. We present preliminary evidence that a prompting-based GPT classifier could be helpful in filtering the generated data to enhance its quality. 1

BackgroundThere is a continual rise in the prevalence of non-cancerous conditions such as chronic kidney disease (CKD) owing to an enormous load of diabetes, hypertension, and vascular diseases. A positive attitude and healthy lifestyle for CKD prevention can only be followed when the masses are well aware and educated about the disease. This study aimed to compare, correlate, and evaluate the distribution of knowledge, attitudes, and perceptions among relatives or caretakers of patients with kidney disease or at risk of the disease. MethodologyThis cross-sectional study aimed at obtaining data on the knowledge, attitudes, and perceptions using the Chronic Kidney Diseases Screening Index questionnaire from the relatives of CKD patients. All data were computed and analyzed using SPSS version 28.0 (IBM Corp., Armonk, NY, USA). ResultsThe majority of the relatives of CKD patients had poor knowledge (63.6%) and poor attitude (51.6%) levels. On the contrary, most respondents had good practices (52.8%) level toward the risk for CKD. A significant correlation was noted between education and knowledge (p < 0.050). A significant association was also observed between education and occupation with attitude (p < 0.001 and p < 0.050, respectively). Additionally, a significant association was noted between age and perception (p < 0.001). ConclusionsInformed and well-educated populations are less prone to acquire or progress to CKD. From this study, we can understand the need for improvement in public knowledge, which has the potential to help in saving the lives of many patients progressing toward end-stage renal diseases.

Effective fusion of data from multiple modalities, such as video, speech, and text, is challenging due to the heterogeneous nature of multimodal data. In this paper, we propose adaptive fusion techniques that aim to model context from different modalities effectively. Instead of defining a deterministic fusion operation, such as concatenation, for the network, we let the network decide "how" to combine a given set of multimodal features more effectively. We propose two networks: 1) Auto-Fusion, which learns to compress information from different modalities while preserving the context, and 2) GAN-Fusion, which regularizes the learned latent space given context from complementing modalities. A quantitative evaluation on the tasks of multimodal machine translation and emotion recognition suggests that our lightweight, adaptive networks can better model context from other modalities than existing methods, many of which employ massive transformer-based networks. 1

This paper presents fast and effective technique for determining the size of pellets fro m real-time v ideo stream. The circular Hough transforms had been used by several researches in iris detection for face recognition, auto matic ball recognition and detecting fingertips position. Here we used circular Hough transform to determine size of pellets in steel plant. This system consists of five steps i.e., take input image, convert into grey scale image, normalize grey scale image, detect edges & perform circular hough transform and do the size analysis. In itially the real-t ime RGB image is being read fro m the webcam, as the image is obtained it is converted to grey scale image following that the light normalization of the image is done, after all the circu lar hough technique is used to detect the pellets and to determine the size of pellets. Since detection and recognition in noisy and cluttered images is challenging problem in co mputer vision. So me of the pellets are overlapping each other and some itself noise that make the recognition process challenging. So to get rid fro m all such difficulties the technique used is circular hough transform. Th is technique helps in determining the centres of the pellets and to measure the radius of pellets and to mark the pellets wh ich are selected. The system is so fast that any change in the input image causes change in the output.

No abstract

Generating relevant responses in a dialog is challenging, and requires not only proper modeling of context in the conversation, but also being able to generate fluent sentences during inference. In this paper, we propose a two-step framework based on generative adversarial nets for generating conditioned responses. Our model first learns a meaningful representation of sentences by autoencoding, and then learns to map an input query to the response representation, which is in turn decoded as a response sentence. Both quantitative and qualitative evaluations show that our model generates more fluent, relevant, and diverse responses than existing state-of-the-art methods. 1

Data augmentation is a widely employed technique to alleviate the problem of data scarcity. In this work, we propose a prompting-based approach to generate labelled training data for intent classification with off-the-shelf language models (LMs) such as GPT-3. An advantage of this method is that no task-specific LM-fine-tuning for data generation is required; hence the method requires no hyper-parameter tuning and is applicable even when the available training data is very scarce. We evaluate the proposed method in a few-shot setting on four diverse intent classification tasks. We find that GPT-generated data significantly boosts the performance of intent classifiers when intents in consideration are sufficiently distinct from each other. In tasks with semantically close intents, we observe that the generated data is less helpful. Our analysis shows that this is because GPT often generates utterances that belong to a closely-related intent instead of the desired one. We present preliminary evidence that a prompting-based GPT classifier could be helpful in filtering the generated data to enhance its quality. 1