Learn More. The Peer-to-Peer request must be received by Maryland Amerigroup maryland prior authorization Care within two 2 business days of the initial notification of the denial. The intent of the Peer-to-Peer is to discuss the denial decision with the ordering clinician or attending physician. For specific details prioe authorization requirements, please refer to our Quick Reference Guide. Certain carefirst mental providers require prior authorization regardless of place of service.
The ETP4G dataset is used to train an empathetic transfer model. Retrieval-based conversational systems learn to rank response candidates for a given dialogue context by computing the similarity between their vector representations. However, training on a single textual form of the multi-turn context limits the ability of a model to learn representations that generalize to natural perturbations seen during inference.
In this paper we propose a framework that incorporates augmented versions of a dialogue context into the learning objective. We utilize contrastive learning as an auxiliary objective to learn robust dialogue context representations that are invariant to perturbations injected through the augmentation method. We experiment with four benchmark dialogue datasets and demonstrate that our framework combines well with existing augmentation methods and can significantly improve over baseline BERT-based ranking architectures.
Furthermore, we propose a novel data augmentation method, ConMix, that adds token level perturbations through stochastic mixing of tokens from other contexts in the batch.
We show that our proposed augmentation method outperforms previous data augmentation approaches, and provides dialogue representations that are more robust to common perturbations seen during inference.
We consider few-shot out-of-distribution OOD intent detection, a practical and important problem for the development of task-oriented dialogue systems.
Despite its importance, this problem is seldom studied in the literature, let alone examined in a systematic way. In this work, we take a closer look at this problem and identify key issues for research. In our pilot study, we reveal the reason why existing OOD intent detection methods are not adequate in dealing with this problem.
Based on the observation, we propose a promising approach to tackle this problem based on latent representation generation and self-supervision. Comprehensive experiments on three real-world intent detection benchmark datasets demonstrate the high effectiveness of our proposed approach and its great potential in improving state-of-the-art methods for few-shot OOD intent detection.
Consistency identification in task-oriented dialog CI-ToD usually consists of three subtasks, aiming to identify inconsistency between current system response and current user response, dialog history and the corresponding knowledge base.
Specifically, CGIM relies on two core insights, referred to as guided multi-head attention module and cycle interactive mechanism, that collaborate from each other. On the one hand, each two tasks are linked with the guided multi-head attention module, aiming to explicitly model the interaction across two related tasks. On the other hand, we further introduce cycle interactive mechanism that focuses on facilitating model to exchange information among the three correlated sub-tasks via a cycle interaction manner.
Experimental results on CI-ToD benchmark show that our model achieves the state-of-the-art performance, pushing the overall score to In addition, we find that CGIM is robust to the initial task flow order. Knowledge-grounded dialog systems need to incorporate smooth transitions among knowledge selected for generating responses, to ensure that dialog flows naturally. For document-grounded dialog systems, the inter- and intra-document knowledge relations can be used to model such conversational flows.
We develop a novel Multi-Document Co-Referential Graph Coref-MDG to effectively capture the inter-document relationships based on commonsense and similarity and the intra-document co-referential structures of knowledge segments within the grounding documents.
CorefDiffs performs knowledge selection by accounting for contextual graph structures and the knowledge difference sequences. CorefDiffs significantly outperforms the state-of-the-art by 9. This demonstrates that the effective modeling of co-reference and knowledge difference for dialog flows are critical for transitions in document-grounded conversation. The core idea is to model the correlation between turn quality and the entire dialogue quality.
We first propose a novel automatic data construction method that can automatically assign fine-grained scores for arbitrarily dialogue data. Then we train SelF-Eval with a multi-level contrastive learning schema which helps to distinguish different score levels. Experimental results on multiple benchmarks show that SelF-Eval is highly consistent with human evaluations and better than the state-of-the-art models. We give a detailed analysis of the experiments in this paper.
Our code is available on GitHub. Automatic evaluation of open-domain dialogs remains an unsolved problem. Existing methods do not correlate strongly with human annotations. In this paper, we present a new automated evaluation method based on the use of follow-ups. We measure the probability that a language model will continue the conversation with a fixed set of follow-ups e.
When compared against twelve existing methods, our new evaluation achieves the highest correlation with human evaluations. The success rate of goals directly correlates with user satisfaction and perceived usefulness of the DS. In this paper, we propose a novel automatic dialogue evaluation framework that jointly performs two tasks: goal segmentation and goal success prediction.
Using an annotated dataset from a commercial DS, we demonstrate that our proposed model reaches an accuracy that is on-par with single-pass human annotation comparing to a three-pass gold annotation benchmark.
In knowledge-grounded dialogue generation, pre-trained language models PLMs can be expected to deepen the fusing of dialogue context and knowledge because of their superior ability of semantic understanding. Unlike adopting the plain text knowledge, it is thorny to leverage the structural commonsense knowledge when using PLMs because most PLMs can only operate plain texts.
Thus, linearizing commonsense knowledge facts into plan text is a compulsory trick. However, a dialogue is always aligned to a lot of retrieved fact candidates; as a result, the linearized text is always lengthy and then significantly increases the burden of using PLMs.
In the first pre-screening stage, we use a ranking network PriorRanking to estimate the relevance of a retrieved knowledge fact. Thus, facts can be clustered into three sections of different priorities. As priority decreases, the relevance decreases, and the number of included facts increases. In the next dialogue generation stage, we use section-aware strategies to encode the linearized knowledge.
The powerful but expensive PLM is only used for a few facts in the higher priority sections, reaching the performance-efficiency balance. Both the automatic and human evaluation demonstrate the superior performance of this work. Personalized response selection systems are generally grounded on persona. However, a correlation exists between persona and empathy, which these systems do not explore well. Also, when a contradictory or off-topic response is selected, faithfulness to the conversation context plunges.
This paper attempts to address these issues by proposing a suite of fusion strategies that capture the interaction between persona, emotion, and entailment information of the utterances. Ablation studies on the Persona-Chat dataset show that incorporating emotion and entailment improves the accuracy of response selection.
We combine our fusion strategies and concept-flow encoding to train a BERT-based model which outperforms the previous methods by margins larger than 2. Dialogue Act tagging with the ISO standard is a difficult task that involves multi-label text classification across a diverse set of labels covering semantic, syntactic and pragmatic aspects of dialogue. The lack of an adequately sized training set annotated with this standard is a major problem when using the standard in practice. In this work we propose a neural architecture to increase classification accuracy, especially on low-frequency fine-grained tags.
Our model takes advantage of the hierarchical structure of the ISO taxonomy and utilises syntactic information in the form of Part-Of-Speech and dependency tags, in addition to contextual information from previous turns.
We train our architecture on an aggregated corpus of conversations from different domains, which provides a variety of dialogue interactions and linguistic registers. Our approach achieves state-of-the-art tagging results on the DialogBank benchmark data set, providing empirical evidence that this architecture can successfully generalise to different domains. Pre-training methods with contrastive learning objectives have shown remarkable success in dialog understanding tasks.
However, current contrastive learning solely considers the self-augmented dialog samples as positive samples and treats all other dialog samples as negative ones, which enforces dissimilar representations even for dialogs that are semantically related. In this paper, we propose SPACE-2, a tree-structured pre-trained conversation model, which learns dialog representations from limited labeled dialogs and large-scale unlabeled dialog corpora via semi-supervised contrastive pre-training.
Concretely, we first define a general semantic tree structure STS to unify the inconsistent annotation schema across different dialog datasets, so that the rich structural information stored in all labeled data can be exploited.
Then we propose a novel multi-view score function to increase the relevance of all possible dialogs that share similar STSs and only push away other completely different dialogs during supervised contrastive pre-training. To fully exploit unlabeled dialogs, a basic self-supervised contrastive loss is also added to refine the learned representations.
Experiments show that our method can achieve new state-of-the-art results on the DialoGLUE benchmark consisting of seven datasets and four popular dialog understanding tasks. Conversational machine reading comprehension CMRC aims to assist computers to understand an natural language text and thereafter engage in a multi-turn conversation to answer questions related to the text. Existing methods typically require three steps: 1 decision making based on entailment reasoning; 2 span extraction if required by the above decision; 3 question rephrasing based on the extracted span.
However, for nearly all these methods, the span extraction and question rephrasing steps cannot fully exploit the fine-grained entailment reasoning information in decision making step because of their relative independence, which will further enlarge the information gap between decision making and question phrasing. Thus, to tackle this problem, we propose a novel end-to-end framework for conversational machine reading comprehension based on shared parameter mechanism, called entailment reasoning T5 ET5.
Despite the lightweight of our proposed framework, experimental results show that the proposed ET5 achieves new state-of-the-art results on the ShARC leaderboard with the BLEU-4 score of Our model and code are publicly available.
Conversational question generation CQG serves as a vital task for machines to assist humans, such as interactive reading comprehension, through conversations. Compared to traditional single-turn question generation SQG , CQG is more challenging in the sense that the generated question is required not only to be meaningful, but also to align with the provided conversation.
Previous studies mainly focus on how to model the flow and alignment of the conversation, but do not thoroughly study which parts of the context and history are necessary for the model. We believe that shortening the context and history is crucial as it can help the model to optimise more on the conversational alignment property. In particular, it selects the top-p sentences and history turns by calculating the relevance scores of them.
Our model achieves state-of-the-art performances on CoQA in both the answer-aware and answer-unaware settings. Pre-trained language models have made great progress on dialogue tasks. However, these models are typically trained on surface dialogue text, thus are proven to be weak in understanding the main semantic meaning of a dialogue context. We investigate Abstract Meaning Representation AMR as explicit semantic knowledge for pre-training models to capture the core semantic information in dialogues during pre-training.
In particular, we propose a semantic-based pre-training framework that extends the standard pre-training framework Devlin et al. Experiments on the understanding of both chit-chats and task-oriented dialogues show the superiority of our model. To our knowledge, we are the first to leverage a deep semantic representation for dialogue pre-training.
Out-of-Domain OOD detection is a key component in a task-oriented dialog system, which aims to identify whether a query falls outside the predefined supported intent set.
Previous softmax-based detection algorithms are proved to be overconfident for OOD samples. Our method is flexible and easily pluggable to existing softmax-based baselines and gains Further analyses show the effectiveness of Bayesian learning for OOD detection.
Due to the increasing use of service chatbots in E-commerce platforms in recent years, customer satisfaction prediction CSP is gaining more and more attention.
CSP is dedicated to evaluating subjective customer satisfaction in conversational service and thus helps improve customer service experience. However, previous methods focus on modeling customer-chatbot interaction across different turns, which are hard to represent the important dynamic satisfaction states throughout the customer journey. In this work, we investigate the problem of satisfaction states tracking and its effects on CSP in E-commerce service chatbots.
In particular, we explore a novel two-step interaction module to represent the dynamic satisfaction states at each turn. In order to capture dialogue-level satisfaction states for CSP, we further introduce dialogue-aware attentions to integrate historical informative cues into the interaction module.
Experiment results demonstrate that our model significantly outperforms multiple baselines, illustrating the benefits of satisfaction states tracking on CSP. Multi-class unknown intent detection has made remarkable progress recently.
However, it has a strong assumption that each utterance has only one intent, which does not conform to reality because utterances often have multiple intents.
In this paper, we propose a more desirable task, multi-label unknown intent detection, to detect whether the utterance contains the unknown intent, in which each utterance may contain multiple intents. In this task, the unique utterances simultaneously containing known and unknown intents make existing multi-class methods easy to fail. To address this issue, we propose an intuitive and effective method to recognize whether All Intents contained in the utterance are Known AIK. If the number of known intents is less than the number of intents, it implies that the utterance also contains unknown intents.
We benchmark AIK over existing methods, and empirical results suggest that our method obtains state-of-the-art performances. For example, on the MultiWOZ 2. Real human conversation data are complicated, heterogeneous, and noisy, from which building open-domain dialogue systems remains a challenging task. In fact, such dialogue data still contains a wealth of information and knowledge, however, they are not fully explored.
In this paper, we show existing open-domain dialogue generation methods that memorize context-response paired data with autoregressive or encode-decode language models underutilize the training data. In particular, we use BERTScore for retrieval, which gives better qualities of the evidence and generation. Experiments over publicly available datasets demonstrate that our method can help models generate better responses, even such training data are usually impressed as low-quality data. Such performance gain is comparable with those improved by enlarging the training set, even better.
We also found that the model performance has a positive correlation with the relevance of the retrieved evidence. Moreover, our method performed well on zero-shot experiments, which indicates that our method can be more robust to real-world data. Building dialogue generation systems in a zero-shot scenario remains a huge challenge, since the typical zero-shot approaches in dialogue generation rely heavily on large-scale pre-trained language generation models such as GPT-3 and T5.
The research on zero-shot dialogue generation without cumbersome language models is limited due to lacking corresponding parallel dialogue corpora. In this paper, we propose a simple but effective Multilingual learning framework for Zero-shot Dialogue Generation dubbed as MulZDG that can effectively transfer knowledge from an English corpus with large-scale training samples to a non-English corpus with zero samples. Besides, MulZDG can be viewed as a multilingual data augmentation method to improve the performance of the resource-rich language.
First, we construct multilingual code-switching dialogue datasets via translation utterances randomly selected from monolingual English datasets. Then we employ MulZDG to train a unified multilingual dialogue model based on the code-switching datasets. The MulZDG can conduct implicit semantic alignment between different languages.
Experiments on DailyDialog and DSTC7 datasets demonstrate that MulZDG not only achieve competitive performance under zero-shot case compared to training with sufficient examples but also greatly improve the performance of the source language.
Prior studies addressing target-oriented conversational tasks lack a crucial notion that has been intensively studied in the context of goal-oriented artificial intelligence agents, namely, planning.
In this study, we propose the task of Target-Guided Open-Domain Conversation Planning TGCP task to evaluate whether neural conversational agents have goal-oriented conversation planning abilities. Using the TGCP task, we investigate the conversation planning abilities of existing retrieval models and recent strong generative models. The experimental results reveal the challenges facing current technology. Since empathy plays a crucial role in increasing social bonding between people, many studies have designed their own dialogue agents to be empathetic using the well-established method of fine-tuning.
However, they do not use prompt-based in-context learning, which has shown powerful performance in various natural language processing NLP tasks, for empathetic dialogue generation. Although several studies have investigated few-shot in-context learning for empathetic dialogue generation, an in-depth analysis of the generation of empathetic dialogue with in-context learning remains unclear, especially in GPT-3 Brown et al.
In this study, we explore whether GPT-3 can generate empathetic dialogues through prompt-based in-context learning in both zero-shot and few-shot settings. We show that GPT-3 achieves competitive performance with Blender 90M, a state-of-the-art dialogue generative model, on both automatic and human evaluation.
Emotion Recognition in Conversation ERC has attracted increasing attention in the affective computing research field. Few works have considered the emotional interactions, which directly reflect the emotional evolution of speakers in the dialogue.
In this work, we propose a novel Dialogue Emotion Interaction Network, DialogueEIN, to explicitly model the intra-speaker, inter-speaker, global and local emotional interactions to respectively simulate the emotional inertia, emotional stimulus, global and local emotional evolution in dialogues. Our codes and models are released. Health coaching helps patients identify and accomplish lifestyle-related goals, effectively improving the control of chronic diseases and mitigating mental health conditions.
However, health coaching is cost-prohibitive due to its highly personalized and labor-intensive nature.
In this paper, we propose to build a dialogue system that converses with the patients, helps them create and accomplish specific goals, and can address their emotions with empathy. However, building such a system is challenging since real-world health coaching datasets are limited and empathy is subtle. Thus, we propose a modularized health coaching dialogue with simplified NLU and NLG frameworks combined with mechanism-conditioned empathetic response generation.
Through automatic and human evaluation, we show that our system generates more empathetic, fluent, and coherent responses and outperforms the state-of-the-art in NLU tasks while requiring less annotation. We view our approach as a key step towards building automated and more accessible health coaching systems. Traditional intent classification models are based on a pre-defined intent set and only recognize limited in-domain IND intent classes.
But users may input out-of-domain OOD queries in a practical dialogue system. Such OOD queries can provide directions for future improvement. We hope to simultaneously classify a set of labeled IND intent classes while discovering and recognizing new unlabeled OOD types incrementally.
We construct three public datasets for different application scenarios and propose two kinds of frameworks, pipeline-based and end-to-end for future work. Further, we conduct exhaustive experiments and qualitative analysis to comprehend key challenges and provide new guidance for future GID research. Medication recommendation is a crucial task for intelligent healthcare systems.
Previous studies mainly recommend medications with electronic health records EHRs. However, some details of interactions between doctors and patients may be ignored or omitted in EHRs, which are essential for automatic medication recommendation. Therefore, we make the first attempt to recommend medications with the conversations between doctors and patients. It contains 11, medical dialogues related to 16 common diseases from 3 departments and 70 corresponding common medications.
Furthermore, we propose a Dialogue structure and Disease knowledge aware Network DDN , where a QA Dialogue Graph mechanism is designed to model the dialogue structure and the knowledge graph is used to introduce external disease knowledge. The extensive experimental results demonstrate that the proposed method is a promising solution to recommend medications with medical dialogues. We propose a speaker clustering model for textual dialogues, which groups the utterances of a multi-party dialogue without speaker annotations, so that the actual speakers are identical inside each cluster.
We find that, without knowing the speakers, the interactions between utterances are still implied in the text, which suggest the relations between speakers. In this work, we model the semantic content of utterance with a pre-trained language model, and the relations between speakers with an utterance-level pairwise matrix.
The semantic content representation can be further instructed by cross-corpus dialogue act modeling. The speaker labels are finally generated by spectral clustering. Experiments show that our model outperforms the sequence classification baseline, and benefits from the auxiliary dialogue act classification task.
We also discuss the detail of determining the number of speakers clusters , eliminating the interference caused by semantic similarity, and the impact of utterance distance. Target-oriented dialog aims to reach a global target through multi-turn conversation. The key to the task is the global planning towards the target, which flexibly guides the dialog concerning the context.
However, existing target-oriented dialog works take a local and greedy strategy for response generation, where global planning is absent. In this work, we propose global planning for target-oriented dialog on a commonsense knowledge graph KG.
We design a global reinforcement learning with the planned paths to flexibly adjust the local response generation model towards the global target. We also propose a KG-based method to collect target-oriented samples automatically from the chit-chat corpus for model training. Experiments show that our method can reach the target with a higher success rate, fewer turns, and more coherent responses.
We examine the link between facets of Rhetorical Structure Theory RST and the selection of content for extractive summarisation, for German-language texts. For this purpose, we produce a set of extractive summaries for a dataset of German-language newspaper commentaries, a corpus which already has several layers of annotation. We provide an in-depth analysis of the connection between summary sentences and several RST-based features and transfer these insights to various automated summarisation models.
Our results show that RST features are informative for the task of extractive summarisation, particularly nuclearity and relations at sentence-level. The state of bridging resolution research is rather unsatisfactory: not only are state-of-the-art resolvers evaluated in unrealistic settings, but the neural models underlying these resolvers are weaker than those used for entity coreference resolution.
In light of these problems, we evaluate bridging resolvers in an end-to-end setting, strengthen them with better encoders, and attempt to gain a better understanding of them via perturbation experiments and a manual analysis of their outputs. Presuppositions are assumptions that are taken for granted by an utterance, and identifying them is key to a pragmatic interpretation of language.
In this paper, we investigate the capabilities of transformer models to perform NLI on cases involving presupposition. Second, to better understand how the model is making its predictions, we analyze samples from sub-datasets of ImpPres and examine model performance on them.
Overall, our findings suggest that NLI-trained transformer models seem to be exploiting specific structural and lexical cues as opposed to performing some kind of pragmatic reasoning.
We present a corrected version of a subset of the FactBank data set. Previously published results on FactBank are no longer valid. We perform experiments on FactBank using multiple training paradigms, data smoothing techniques, and polarity classifiers. We argue that f-measure is an important alternative evaluation metric for factuality.
We provide new state-of-the-art results for four corpora including FactBank. We perform an error analysis on Factbank combined with two similar corpora. Discourse parsing has proven to be useful for a number of NLP tasks that require complex reasoning. However, over a decade since the advent of the Penn Discourse Treebank, predicting implicit discourse relations in text remains challenging. There are several possible reasons for this, and we hypothesize that models should be exposed to more context as it plays an important role in accurate human annotation; meanwhile adding uncertainty measures can improve model accuracy and calibration.
To thoroughly investigate this phenomenon, we perform a series of experiments to determine 1 the effects of context on human judgments, and 2 the effect of quantifying uncertainty with annotator confidence ratings on model accuracy and calibration which we measure using the Brier score Brier et al, We also find some insightful qualitative results regarding human and model behavior on these datasets.
Contingent reasoning is one of the essential abilities in natural language understanding, and many language resources annotated with contingent relations have been constructed. However, despite the recent advances in deep learning, the task of contingent reasoning is still difficult for computers. In this study, we focus on the reasoning of contingent relation between basic events. Based on the existing data construction method, we automatically generate large-scale pseudo-problems and incorporate the generated data into training.
We also investigate the generality of contingent knowledge through quantitative evaluation by performing transfer learning on the related tasks: discourse relation analysis, the Japanese Winograd Schema Challenge, and the JCommonsenseQA. The experimental results show the effectiveness of utilizing pseudo-problems for both the commonsense contingent reasoning task and the related tasks, which suggests the importance of contingent reasoning.
Irony is a ubiquitous figurative language in daily communication. Previously, many researchers have approached irony from linguistic, cognitive science, and computational aspects. Recently, some progress have been witnessed in automatic irony processing due to the rapid development in deep neural models in natural language processing NLP.
In this paper, we will provide a comprehensive overview of computational irony, insights from linguisic theory and cognitive science, as well as its interactions with downstream NLP tasks and newly proposed multi-X irony processing perspectives.
We compare the performance of a pattern-based approach and a sequence labeling model, add an experiment on the pre-classification of candidate sentences, and provide an initial qualitative analysis of the error cases made by both models.
Humans use different wordings depending on the context to facilitate efficient communication. For example, instead of completely new information, information related to the preceding context is typically placed at the sentence-initial position. In this study, we analyze whether neural language models LMs can capture such discourse-level preferences in text generation.
Specifically, we focus on a particular aspect of discourse, namely the topic-comment structure. To analyze the linguistic knowledge of LMs separately, we chose the Japanese language, a topic-prominent language, for designing probing tasks, and we created human topicalization judgment data by crowdsourcing. Our experimental results suggest that LMs have different generalizations from humans; LMs exhibited less context-dependent behaviors toward topicalization judgment.
These results highlight the need for the additional inductive biases to guide LMs to achieve successful discourse-level generalization. A critical component of competence in language is being able to identify relevant components of an utterance and reply appropriately.
In this paper we examine the extent of such dialogue response sensitivity in pre-trained language models, conducting a series of experiments with a particular focus on sensitivity to dynamics involving phenomena of at-issueness and ellipsis. We find that models show clear sensitivity to a distinctive role of embedded clauses, and a general preference for responses that target main clause content of prior utterances.
However, the results indicate mixed and generally weak trends with respect to capturing the full range of dynamics involved in targeting at-issue versus not-at-issue content. Additionally, models show fundamental limitations in grasp of the dynamics governing ellipsis, and response selections show clear interference from superficial factors that outweigh the influence of principled discourse constraints. Recent research shows that pre-trained language models, built to generate text conditioned on some context, learn to encode syntactic knowledge to a certain degree.
This has motivated researchers to move beyond the sentence-level and look into their ability to encode less studied discourse-level phenomena. In this paper, we add to the body of probing research by investigating discourse entity representations in large pre-trained language models in English. Motivated by early theories of discourse and key pieces of previous work, we focus on the information-status of entities as discourse-new or discourse-old.
We present two probing models, one based on binary classification and another one on sequence labeling. The results of our experiments show that pre-trained language models do encode information on whether an entity has been introduced before or not in the discourse.
However, this information alone is not sufficient to find the entities in a discourse, opening up interesting questions about the definition of entities for future work. While neural approaches to argument mining AM have advanced considerably, most of the recent work has been limited to parsing monologues.
With an urgent interest in the use of conversational agents for broader societal applications, there is a need to advance the state-of-the-art in argument parsers for dialogues.
This enables progress towards more purposeful conversations involving persuasion, debate and deliberation. This paper discusses Dialo-AP, an end-to-end argument parser that constructs argument graphs from dialogues. We formulate AM as dependency parsing of elementary and argumentative discourse units; the system is trained using extensive pre-training and curriculum learning comprising nine diverse corpora.
Dialo-AP is capable of generating argument graphs from dialogues by performing all sub-tasks of AM. Compared to existing state-of-the-art baselines, Dialo-AP achieves significant improvements across all tasks, which is further validated through rigorous human evaluation.
Implicit Discourse Relation Recognition IDRR is to detect and classify relation sense between two text segments without an explicit connective. Vanilla pre-train and fine-tuning paradigm builds upon a Pre-trained Language Model PLM with a task-specific neural network.
However, the task objective functions are often not in accordance with that of the PLM. Furthermore, this paradigm cannot well exploit some linguistic evidence embedded in the pre-training process. The recent pre-train, prompt, and predict paradigm selects appropriate prompts to reformulate downstream tasks, so as to utilizing the PLM itself for prediction. However, for its success applications, prompts, verbalizer as well as model training should still be carefully designed for different tasks.
As the first trial of using this new paradigm for IDRR, this paper develops a Connective-cloze Prompt ConnPrompt to transform the relation prediction task as a connective-cloze task. Specifically, we design two styles of ConnPrompt template: Insert-cloze Prompt ICP and Prefix-cloze Prompt PCP and construct an answer space mapping to the relation senses based on the hierarchy sense tags and implicit connectives.
Furthermore, we use a multi-prompt ensemble to fuse predictions from different prompting results. Experiments on the PDTB corpus show that our method significantly outperforms the state-of-the-art algorithms, even with fewer training data. Conversational discourse parsing aims to construct an implicit utterance dependency tree to reflect the turn-taking in a multi-party conversation.
Existing works are generally divided into two lines: graph-based and transition-based paradigms, which perform well for short-distance and long-distance dependency links, respectively. However, there is no study to consider the advantages of both paradigms to facilitate conversational discourse parsing.
As a result, we propose a distance-aware multi-task framework DAMT that incorporates the strengths of transition-based paradigm to facilitate the graph-based paradigm from the encoding and decoding process. To promote multi-task learning on two paradigms, we first introduce an Encoding Interactive Module EIM to enhance the flow of semantic information between both two paradigms during the encoding step. And then we apply a Distance-Aware Graph Convolutional Network DAGCN in the decoding process, which can incorporate the different-distance dependency links predicted by the transition-based paradigm to facilitate the decoding of the graph-based paradigm.
The experimental results on the datasets STAC and Molweni show that our method can significantly improve the performance of the SOTA graph-based paradigm on long-distance dependency links. This work deploys linguistically motivated features to classify paragraph-level text into fiction and non-fiction genre using a logistic regression model and infers lexical and syntactic properties that distinguish the two genres.
Previous works have focused on classifying document-level text into fiction and non-fiction genres, while in this work, we deal with shorter texts which are closer to real-world applications like sentiment analysis of tweets.
Going beyond simple POS tag ratios proposed in Qureshi et al. For the task of short-text classification, a model containing 28 best-features selected via Recursive feature elimination with cross-validation; RFECV confers an accuracy jump of The efficacy of the above model containing a linguistically motivated feature set also transfers over to another dataset viz, Baby BNC corpus. We also compared the classification accuracy of the logistic regression model with two deep-learning models.
Although both the deep learning models give better results in terms of classification accuracy, the problem of interpreting these models remains unsolved.
In contrast, regression model coefficients revealed that fiction texts tend to have more character-level diversity and have lower lexical density quantified using content-function word ratios compared to non-fiction texts. Moreover, subtle differences in word order exist between the two genres, i. Though successes have been observed, embedding the whole syntactic structures as one vector inevitably overlooks the fine-grained syntax matching patterns, e. In this paper, we formalize the task of semantic sentence matching as a problem of graph matching in which each sentence is represented as a directed graph according to its syntactic structures.
The syntax matching patterns i. After that, the neural quadratic assignment programming QAP is adapted to extract syntactic matching patterns from the association graph. In this way, the syntactic structures fully interact in a fine granularity during the matching process. Experimental results on three public datasets demonstrated that ISG can outperform the state-of-the-art baselines effectively and efficiently. The empirical analysis also showed that ISG can match sentences in an interpretable way.
Text classification is a primary task in natural language processing NLP. Recently, graph neural networks GNNs have developed rapidly and been applied to text classification tasks. As a special kind of graph data, the tree has a simpler data structure and can provide rich hierarchical information for text classification. Inspired by the structural entropy, we construct the coding tree of the graph by minimizing the structural entropy and propose HINT, which aims to make full use of the hierarchical information contained in the text for the task of text classification.
Specifically, we first establish a dependency parsing graph for each text. Then we designed a structural entropy minimization algorithm to decode the key information in the graph and convert each graph to its corresponding coding tree. Based on the hierarchical structure of the coding tree, the representation of the entire graph is obtained by updating the representation of non-leaf nodes in the coding tree layer by layer. Finally, we present the effectiveness of hierarchical information in text classification.
Experimental results show that HINT outperforms the state-of-the-art methods on popular benchmarks while having a simple structure and few parameters. The conventional success of textual classification relies on annotated data, and the new paradigm of pre-trained language models PLMs still requires a few labeled data for downstream tasks. However, in real-world applications, label noise inevitably exists in training data, damaging the effectiveness, robustness, and generalization of the models constructed on such data.
Recently, remarkable achievements have been made to mitigate this dilemma in visual data, while only a few explore textual data. To fill this gap, we present SelfMix, a simple yet effective method, to handle label noise in text classification tasks. SelfMix uses the Gaussian Mixture Model to separate samples and leverages semi-supervised learning. Unlike previous works requiring multiple models, our method utilizes the dropout mechanism on a single model to reduce the confirmation bias in self-training and introduces a textual level mixup training strategy.
Experimental results on three text classification benchmarks with different types of text show that the performance of our proposed method outperforms these strong baselines designed for both textual and visual data under different noise ratios and noise types. We present our novel, hyperparameter-free topic modelling algorithm, Community Topic.
Our algorithm is based on mining communities from term co-occurrence networks. We empirically evaluate and compare Community Topic with Latent Dirichlet Allocation and the recently developed top2vec algorithm. We find that Community Topic runs faster than the competitors and produces topics that achieve higher coherence scores. Community Topic can discover coherent topics at various scales. The network representation used by Community Topic results in a natural relationship between topics and a topic hierarchy.
This allows sub- and super-topics to be found on demand. These features make Community Topic the ideal tool for downstream applications such as applied research and conversational agents. Nowadays, deep-learning based NLP models are usually trained with large-scale third-party data which can be easily injected with malicious backdoors. Text-based BDA aims to train a poisoned model with both clean and poisoned texts to perform normally on clean inputs while being misled to predict those trigger-embedded texts as target labels set by attackers.
Previous works usually choose fixed Positions-to-Poison P2P first, then add triggers upon those positions such as letter insertion or deletion. However, considering the positions of words with important semantics may vary in different contexts, fixed P2P models are severely limited in flexibility and performance.
We study the text-based BDA from the perspective of automatically and dynamically selecting P2P from contexts. We design a novel Locator model which can predict P2P dynamically without human intervention. Experiments on two public datasets show both tinier test accuracy gap on clean data and higher attack success rate on poisoned ones.
Human evaluation with volunteers also shows the P2P predicted by our model are important for classification. Explaining the predictions of a deep neural network DNN is a challenging problem. Many attempts at interpreting those predictions have focused on attribution-based methods, which assess the contributions of individual features to each model prediction.
However, attribution-based explanations do not always provide faithful explanations to the target model, e. We present a method to learn explanations-specific representations while constructing deep network models for text classification.
These representations can be used to faithfully interpret black-box predictions, i. We show that learning specific representations improves model interpretability across various tasks, for both qualitative and quantitative evaluations, while preserving predictive performance. We address contextualized code retrieval, the search for code snippets helpful to fill gaps in a partial input program. Our approach facilitates a large-scale self-supervised contrastive training by splitting source code randomly into contexts and targets.
To combat leakage between the two, we suggest a novel approach based on mutual identifier masking, dedentation, and the selection of syntax-aligned targets.
Our second contribution is a new dataset for direct evaluation of contextualized code retrieval, based on a dataset of manually aligned subpassages of code clones. Our experiments demonstrate that the proposed approach improves retrieval substantially, and yields new state-of-the-art results for code clone and defect detection.
We present a biomedical knowledge enhanced pre-trained language model for medicinal product vertical search. Furthermore, we propose a novel pre-training task, product attribute prediction PAP , to inject product knowledge into the pre-trained language model efficiently by leveraging medicinal product databases directly.
Experiments demonstrate the effectiveness of PAP task for pre-trained language model on medicinal product vertical search scenario, which includes query-title relevance, query intent classification, and named entity recognition in query.
Fact-checking has gained increasing attention due to the widespread of falsified information. Most fact-checking approaches focus on claims made in English only due to the data scarcity issue in other languages. The lack of fact-checking datasets in low-resource languages calls for an effective cross-lingual transfer technique for fact-checking. Additionally, trustworthy information in different languages can be complementary and helpful in verifying facts.
To this end, we present the first fact-checking framework augmented with cross-lingual retrieval that aggregates evidence retrieved from multiple languages through a cross-lingual retriever.
Given the absence of cross-lingual information retrieval datasets with claim-like queries, we train the retriever with our proposed Cross-lingual Inverse Cloze Task X-ICT , a self-supervised algorithm that creates training instances by translating the title of a passage.
The goal for X-ICT is to learn cross-lingual retrieval in which the model learns to identify the passage corresponding to a given translated title. On the X-Fact dataset, our approach achieves 2. Enhancing the interpretability of text classification models can help increase the reliability of these models in real-world applications. Currently, most researchers focus on extracting task-specific words from inputs to improve the interpretability of the model.
The competitive approaches exploit the Variational Information Bottleneck VIB to improve the performance of word masking at the word embedding layer to obtain task-specific words.
However, these approaches ignore the multi-level semantics of the text, which can impair the interpretability of the model, and do not consider the risk of representation overlap caused by the VIB, which can impair the classification performance.
In this paper, we propose an enhanced variational word masks approach, named E-VarM, to solve these two issues effectively. The E-VarM combines multi-level semantics from all hidden layers of the model to mask out task-irrelevant words and uses contrastive learning to readjust the distances between representations. Empirical studies on ten benchmark text classification datasets demonstrate that our approach outperforms the SOTA methods in simultaneously improving the interpretability and accuracy of the model.
Metadata attributes e. However, recent models rely on pretrained language models PLMs , in which previously used techniques for attribute injection are either nontrivial or cost-ineffective. In this paper, we introduce a benchmark for evaluating attribute injection models, which comprises eight datasets across a diverse range of tasks and domains and six synthetically sparsified ones.
We also propose a lightweight and memory-efficient method to inject attributes into PLMs. We extend adapters, i. We use approximation techniques to parameterize the model efficiently for domains with large attribute vocabularies, and training mechanisms to handle multi-labeled and sparse attributes.
Extensive experiments and analyses show that our method outperforms previous attribute injection methods and achieves state-of-the-art performance on all datasets. Research on neural IR has so far been focused primarily on standard supervised learning settings, where it outperforms traditional term matching baselines.
Many practical use cases of such models, however, may involve previously unseen target domains. In this paper, we propose to improve the out-of-domain generalization of Dense Passage Retrieval DPR - a popular choice for neural IR - through synthetic data augmentation only in the source domain. We empirically show that pre-finetuning DPR with additional synthetic data in its source domain Wikipedia , which we generate using a fine-tuned sequence-to-sequence generator, can be a low-cost yet effective first step towards its generalization.
Across five different test sets, our augmented model shows more robust performance than DPR in both in-domain and zero-shot out-of-domain evaluation.
State-of-the-art neural re rankers are notoriously data-hungry which given the lack of large-scale training data in languages other than English makes them rarely used in multilingual and cross-lingual retrieval settings. Current approaches therefore commonly transfer rankers trained on English data to other languages and cross-lingual setups by means of multilingual encoders: they fine-tune all parameters of pretrained massively multilingual Transformers MMTs, e.
In this work, we show that two parameter-efficient approaches to cross-lingual transfer, namely Sparse Fine-Tuning Masks SFTMs and Adapters, allow for a more lightweight and more effective zero-shot transfer to multilingual and cross-lingual retrieval tasks.
At inference, this modular design allows us to compose the ranker by applying the re ranking adapter or SFTM trained with source language data together with the language adapter or SFTM of a target language. We carry out a large scale evaluation on the CLEF and HC4 benchmarks and additionally, as another contribution, extend the former with queries in three new languages: Kyrgyz, Uyghur and Turkish.
The proposed parameter-efficient methods outperform standard zero-shot transfer with full MMT fine-tuning, while being more modular and reducing training times. The gains are particularly pronounced for low-resource languages, where our approaches also substantially outperform the competitive machine translation-based rankers. In weakly-supervised text classification, only label names act as sources of supervision. Predominant approaches to weakly-supervised text classification utilize a two-phase framework, where test samples are first assigned pseudo-labels and are then used to train a neural text classifier.
In most previous work, the pseudo-labeling step is dependent on obtaining seed words that best capture the relevance of each class label. We present LIME, a framework for weakly-supervised text classification that entirely replaces the brittle seed-word generation process with entailment-based pseudo-classification. We find that combining weakly-supervised classification and textual entailment mitigates shortcomings of both, resulting in a more streamlined and effective classification pipeline.
So you have to be on your toes all the time. Do you think technology will be the ultimate weapon to overcome these challenges? We need to remember that we are living in the 21st century age where everything is driven by technology. Earlier people have to go to the branch where they were connected to the technology through PCs and laptops but now the range and technology barriers are gone. The Misys Solution gave us an advantage to uniquely adapt the solution to our requirements and not restrict us only to the standard out of the box functionalities with restrictions on security, workflow and the usual challenges that a Bank usually faces.
Customer Linked Finance and cash flow rationalization have always been a priority for all customers whether big or small. With Indian business touching new heights,transaction banking has come to the forefront in the last couple of years. The services may differ for ex-Trade Finance, Cash Management, Supply Chain Finance, Treasury or Lending Services STP has always been the main focus to enable to all our corporate to get a synchronous view of their positions and ability to bank with us round the clock without any manual intervention.
Significance of Technology in New-age Co-operative Banks Co-operative banks have traditionally played an integral role in helping rural and urban people with credit facilities, especially the ones with no or little access to finances to fund their basic needs. They provided economic security to small businesses and poor segment of the society and gained huge popularity in its initial few decades. However, with increased globalization and intense competition from commercial and private banks, co-operative banks have witnessed a slow pace of growth in the past years.
Government interference, mismanagement, lack of awareness among people, restricted coverage and reluctance to adopt new and efficient technologies are few of the major impediments to their growth.
In the view of enhanced competition and the fact that most of the banking commodities are undifferentiated products, it is customer service that becomes the sole differentiator factor to stay ahead in the business. Banks need to gear up for providing more efficient and cost-effective services leveraging the technological capabilities. Ameyo, a contact center platform and a market leader in omnichannel customer experience, can drive customer experience revolution and transform the way banks communicate with clients, thereby increasing revenues.
Omnichannel banking being the buzzword today, banks have to be capable of delivering seamless customer experience over all devices in order to gain new customers and maintain the existing customers. Importance of Delivering Proactive Customer Service Due to huge penetration of mobile devices and an ongoing shift in customer demographics, banks need to offer proactive care to increase brand loyalty and advocacy as well as to reduce call center inbound call volume.
Proactively reaching out to customers via automated voice, email or text for reminders about bill payments and card balances, how to manage finances or timely news on service updates, creates more than good. IVR- A Blend of Self-Service and Human Support As banks receive a large number of calls every day, answering all of them with the exact information is highly exhaustible and time-consuming.
It enables self-service and simple issues can be easily dealt with it. Voice Blaster to Expand Reach Ameyo offers innovative Voice Broadcasting solution for mass communication and makes it easy for banks to initiate and build relationships with the customers at lower costs.
Voice broadcasting allows the user to send hundreds and thousands of pre-recorded voice messages instantly and simultaneously. Predictive dialing software and automatic dialer algorithms can improve call efficiency by lowering the wait time, number of dropped calls and idle time of agents. Enhanced Customer Experience with Social Media Banks should also take social media seriously and get engaged with its customers on a 24X7 basis. Being active on social media and carefully listening to customers will showcase other potential customers that your bank actually cares and is committed to providing exceptional customer service.
The role of technology in reviving and revolutionizing co-operative banking sector is inevitable. The cooperative Banks must hasten adoption of NPCI product and services, which are expected as a minimum level of service by the customers, says Navneet Kumar of National Payments Corporation of India NPCI in an interview with Paulomi Chakraborty of Elets News Network ENN would increase but something as ground breaking as this would need a lot of treading over customer learning curve as well for the rural banking sector and their customers alike.
While many cooperative banks have expressed interest in the same, most of them are having footprint in the urban landscape TJSB Sahkari Bank was, in fact, part of the pilot and has gone live now. It has done exceptionally well for themselves among the mainline commercial banks. IT has been adopted by entire banking industry. To what extent it has been adopted in the rural banking sector? What role is the NPCI playing in rural banks and financial institutions? This can be looked at from two perspectives the member banks and financial institutions and their customers i.
How is UPI benefitting the rural banking sector? Adoption is on the rise and transactions. A great deal. With NPCI providing level-playing field in terms of systems plus processes, security, charges, etc and good support from the Application Service Providers for the ecosystem reducing the burden on the banks, the technology adoption by the rural banking sector is near whole.
IFTAS is another entity that is doing focussed work for this segment, which will benefit technology adoption further. What steps do you think the rural banking or cooperative banking sector should adopt to bring them at par with the entire banking sector?
Apart from the internal steps mentioned earlier, the cooperative banks must hasten adoption of NPCI product and services, which are expected as a minimum level of service by the customers, even adopt aspiration products that their customer would choose from other banks. Innovate around these product and services to offer custom solutions to another sector for eg.
MFIs we are in the process of facilitating certain pilots. They must recognise that some of the activities like mobile seeding and Aadhaar seeding is not done from compliance perspective but as a potential revenue source. They must also invest in customer education on products and also push for basic financial literacy. In-Solutions Global Pvt. Ltd is the leader in electronic financial transaction domain. Launched in , the company deals with 25 million transactions on a daily basis.
Being the trendsetter in payment domain industry our competency lies in nurturing human capital and building technologically advanced solutions. On a day-to-day basis, the operational complexities of the transactions as well as the entities who are party to the payment card environment have to be provided with a proactive resolution.
The out-of-box thinking and ideas are converted into action. Our solutions are automated based on the subject matter experts and. We not only work on client requirement but also give insights on value adds for next generation consumers. Your card management system has been deployed in 20 banks in last 15 years.
How do you manage end-to-end cycles of credit, debit and pre-paid cards? We understand the end-to-end card lifecycle from source to destination. The entire technology from issuance, dispatch, reconciliation, risk management, dispute management, e-kyc all are in house managed services.
The gamut of product under one single brand ISG helps us to serve in a seamless manner. Your merchant management system has the ability to handle huge transactions, please explain how your 24X7 technical team gives support in dispute management and risk evaluation? ISG- merchant management system handles huge volume of transaction. Our built-in fraud and risk management modules with pre-defined rule engine help to identify risky transaction and filter it.
The extended services like Merchant Sales Pricing, Automated SOC, Sales force automation gives leverage in reduction of disputes and provide overview on business scenarios which are profit to the merchant. The merchant can log on to the portal in case of support; a ticket is raised to respective. The extended services empower the merchants for decision making. How is it tailor-made to cater to the specific needs of the consumer? The modules can be customised depending on the requirement of different bank types.
Thus, any services required can be hosted by ISG. How does ISGPay Payment gateway facilitate the transfer of information between a payment portal such as a website, mobile phone or interactive voice response service and a processor?
ISGpay facilitates payment collection from various sources like debit card, credit card, IMPS, UPI etc for websites, mobile phones or interactive voice response service and a processor. It does payment collection online as well as offline, Offline with POS. ISGpay does risk management, customer analytics, payment analytics for the merchant. It also does intelligent routing of transactions based on success ratio. The objective of this portal is to empower consumers and be aware of the rights which they avail.
The consumers can file cases, store case related documents and track cases with alerts on phone and email. An integrated platform for case filing and court fee payment is boon to the end consumer. Our built in fraud and risk management modules with predefined rule engine help to identify risky transaction and filter it. What are the innovative solutions you have in promoting Digital Payments in the country?
EMV- Smart chip card to run multi applications on the same card will definitely increase the customer base in digital space. With an increased freedom in the digital payment options, more and more focus is on the cyber security. And, one company which has pioneered this field is Avenues India. We, at CCAvenue, have always felt that the best way to improve customer trust in online payments is by offering greater control over the transaction process between merchants and their customers.
PCI DSS aims to ensure that all organisations that process, store or transmit credit card information maintain a secure environment. Besides, we have also achieved ISO certification that not only validates our efforts towards complying with the highest level of security standards, but also reinforces the trust our merchants have placed on us. Tell us about your multiple currency process.
Being cognizant to the evolving needs of our merchant base, CCAvenue has always strived to upgrade its platform to enhance their business potential. We soon realised that a majority of merchants cater to international customers too. But due to lack of currency options during checkout, they miss out on valuable business opportunities. Our team developed a new payment processing platform to meet this requirement.
We introduced multi- currency payment platform so as to make it possible for customers to pay in their local currency and avoid conversion disputes. DCC is available for up to currencies. CCAvenue has partnered with various banks like American Express, Standard Chartered bank for unique banking services, could you highlight some of these services?
Our business relationship with American Express goes back a long way. Even today, CCAvenue constantly forges alliances with leading as well as emerging banks to provide merchants with unique payment options.
Built upon the IMPS mobile payment system, UPI will help our merchants by facilitating instant fund transfers through a unique virtual payment address. How is CCAvenue helping in the e-governance initiative of the government? CCAvenue has always supported government initiatives by enabling payment processing for their products and services. We help several government organisations in accepting online payments for challans, taxes, and other bill payments.
An RBI committee has estimated that over 30, million bills amounting to Rs 6, billion are generated each year in the top 20 cities in the country. CCAvenue will now be able to operate as an individual bill payment collection operating unit. CCAvenue helps several government organisations in accepting online payments for challans, taxes, and other bill payments.
How does CCAvenue help in maintaining the e-wallets of its customers? Today, every other player in the market is coming out with their own wallet as a customer acquisition strategy. Most of them offer discounts, cashbacks and other incentives to encourage higher transactions from customers.
However, CCAvenue has always remained a neutral payment gateway. As more and more wallets are introduced, you can be sure that CCAvenue will include them in its widest payment network. With customers switching to mobile banking over traditional banking, Empays has taken a lead in launching such services for banks. Empays has consistently innovated in the area of mobile financial services offered through banks. This arises from a belief that regulated financial institutions are the key to building and retaining consumer confidence in a turbulent world.
Anyone can receive a payment from an account holder in member banks using just the mobile number and a code. Using this, the person can get cash from any of these ATMs without using any card.
Certified as a payment system by the RBI, the system is one of its kind in the country. As part of the platform, an account or customer relationship can be set up digitally in less than a minute in selected countries outside India. The platform embeds a payment mediation service that meets all the objectives of SWIFT but does so instantly.
Other banks in other countries are in the pipeline. Your instant money transfer is very popular. How does it help the common man? It is the only such system of its kind regulated as a payment system by the Central Bank.
The payment can only be initiated from a bank channel like kiosk, ATM, mobile and can only be withdrawn from a bank channel ATM or Agent. As a result IMT is the cheapest mode of sending money. The payments are also instant. It means there is no question of waiting. Finally, the system is regulated by the Reserve Bank which promises the guarantee of safety that we provide as a result of the supervision by the RBI. Empays payment cloud is all about digital payments.
How does it allow rapid innovation with minimum investment? The APIs that are part of the Empays Payment Cloud are offered as a service that the bank can consume as part of digital payments. The system is hosted on private cloud in a safe and secure manner. A white labelled App is also available which contains the digital flow of payments. All this means the bank can prototype a payment in a month and go live in three to four months with no investment other than a setup fee and transaction-based payment scheme.
How are your products helping in the e-governance initiative of the Government of India? Getting cash and benefits in the hands of people is a massive priority and we are talking to government departments on how best to deliver Direct Benefit.
What are some new products that you plan to launch soon? We will be launching new products that bring social media and banking together in an innovative manner. We are also watching the space of contactless payments very closely and will be present in this space as well. Empays is keen that the government uses native Indian innovations like Instant Money Transfer IMT to handle their payments and not just channels made available by the public sector agencies.
As an alternate payment system, we request a levelplaying field. Secondly, we are keen that the government uses native Indian innovations like IMT to handle their payments and not just channels made available by the public sector agencies. If the government is keen on entrepreneurship, they must reward domestic entrepreneurs with volumes.
Expanding its Horizon Talking about the IT in payments, the whole world is getting transformed due to IT developments in day-today operations. So we were focused on micro customers and we were always focused on providing all the financial services to the customers. There are lots of undermined customers that still exist. These are customers who have balances but not enough to excite the large banks and they are underserved today.
Our focus has always been in this segment because we believe it can be served efficiently and economically well with the multi product offering. So we will continue this journey. As a payments bank, we will provide all four financial products i. How do you pursue development within the IT sector? Talking about the IT in payments, the whole world is getting transformed due to IT developments in dayto-day operations.
One of the key things is people are looking more at the end product and how the customer is benefitting from the technology that he is putting up rather than the technology itself.
So mobile is a big conversation, self service is a big conversation and instant gratification is a big conversation. Banks are thinking out of the box and using everything else ensuring that the customer gets the service at all points of time.
How do you rate Indian banks and other financial institutions as in their capabilities and willingness to adopt new technologies as compared to other western countries? The technology being used in banks in India is at par if not ahead of others. Some of the new entrants are taking the technology to the next level and looking at building a very different international platform than what it has been in the past. Everything in the past was lending-oriented whereas now people are saying that lending is only one part of the business and transactions are the bigger part of the business and let me build something which focuses and delivers the transactions.
Everything else is a by-product. People are looking at the transactions rather than just the loans and accounts. It has taken care of two things. It has taken tokenlisation to the next level. You can make any ally or an acronym to make a payment. Anyone can originate a transaction from the mobile, anyone can accept and pay for that transaction without having to worry about the security or their connections getting compromised.
If right amount of incentives are provided, say in terms of lower interchange for the merchant and maybe some incentive for using UPI for the customers, this can take India to the next level. We generally view customers into two segments --urban and rural. What strategy do you have for each one of these? I am trying to catch up both legs of customers i. We call them mass market, which are customers who earn 3 to 5 lakhs annually. There have been a lot of IT initiatives to launch these operations.
Kindly share about them as well. We started our journey as a technology company to provide technology to the other people. So technology is the core and it continues to be the first brick which we lay in launching any operation. We were the pioneers in biometrics and smart cards in India. It has become mainstream now. We continue to focus on biometrics because we believe that it is an enabler for the customer segment that we work in and we continue to invest in Aadhar.
We are investing a lot in mobile banking because our view is that while the customer may start his day today, with us assisting him in the transaction. We are following fiscal and digital. We will add some branches which will be more like hubs, which will be serving the customers. We are also looking at providing self service terminals and free wi-fi at our branches so that customer can route the connectivity and try and experience the transaction himself. The whole concept of independently being able to consume this service and instant privacy are the things which I think people will enjoy and gradually migrate.
We are starting with branches going to 1, branches and spread over 8 to 9 states and that is the plan. MindCraft Shaping History in Digital World MindCraft, which is a premier industry vertical specific company, provides business edge to clients through world class technology products and services. Today, banks are inundated with the need to innovate on account of global trends, regulatory changes and stakeholder demands. These growing pressures have made it imperative for banks to constantly evolve.
MindCraft provides products and solutions that enable banks to manage huge business volumes, monitor risk, improve TAT etc. PayCraft is one of the first products we had developed. It is a highly scalable platform for processing bulk and online remittance transactions received from Exchange Houses across the globe.
PayConnect is an integrated payment hub that helps banks keep track of all their domestic payment transactions. It provides real-time tracking of all payments processed. Aurum is a loan origination system that automates the entire lending cycle of the loan process. It also includes contractual reporting obligations, process modelling, and automated reports among other features. Another solution we have is MetRisk, which is a risk monitoring solution that delivers the risk exposure of an organization at an Enterprise level.
Developed primarily for the insurance and mutual fund industry, the solution is also available for various types of treasury operations. Kindly tell us about your PayCraft and SalesCtrl solutions. Given the complexities in the remittance business, PayCraft is a solution that we see a lot of demand for. In their quest to be better prepared to manage risk and improve forecasting, we are also seeing an increasing demand for services around Business Analytics and BPM.
With regards to our remittance solution, PayCraft has ready out-of-the box support for over 42 exchange houses in the GCC Gulf Cooperation Council region.
This solution can help banks not only reduce costs related to foreign remittance but it can also help them increase their customer base. SalesCtrl, on the other hand, is a cross industry product developed by us. It is an opportunity management tool that captures and monitors the right information, from identification to closure and can be used by companies, irrespective of the industry vertical to which they belong.
It has been developed primarily for the SME segment especially for companies in the services industry to help them bring about efficiency in their sales process. Can you share names of some of your prominent clients? What feedback have you received from them? We are proud of the long-term relationship we share with our clients. We are often considered to be the right-sized partner by our clients. We are large enough to provide a wide range of solutions and services that cater to the complete IT requirements of our customers.
At the same time, we are small enough to give executive attention to each client engagement. Our clients have often pointed out that our speed to respond and the flexibility to adapt to change makes us a dependable and scalable partner.
What are the main challenges being observed in the Indian banking and finance sector on the technological front? Banks adopted technology very early, about 20 years ago. Initially, the focus was on branch automation. Later, it changed to Core Banking. After this, the development and adoption of technology by the RBI was so fast that banks found it very difficult to cope up with. Hence, most banks in India have a huge number of applications. Large Multinational Banks, mainly in the US, are looking at this option today.
The fact is that all financial services companies, including banks, are finding it increasingly difficult to justify IT costs. As a first step, all secondary data could move out to service providers. In the long run, most businesses, including financial services, will have IT providers who will manage the entire show for a cost per transaction.
What is its significance? We have been working closely with IBM for over 12 years. These days, the game is changing a bit. We need the solution before the customer will consider buying licenses. We have adapted to this change and helped IBM sell their platform to a variety of customers.
Due to our solutions, it has been easier for us to consistently renew the licenses. IBM has always been very keen to ensure that the customers actually use their products and generate value for their organisations. We have been instrumental in helping IBM achieve this goal.
So, there is a lot of promise and lot of hype as well. The reality is yet to be realised. But there are a lot of interests and lots of news articles. Banks and CTOs are also talking about it. How do you define Blockchain technology?
How it is going to help or already helping the banking sector? Blockchain is a new technology. It helps maintaining distribute ledger of transactions that are immutable or final. No one can change the transactions unilaterally.
Everyone has to go through what we call a process of consensus to agree that the transaction is valid and gets added to the code of transactions. That really helps many institutes including the financial industry. So, we have web applications in banking, finance, insurance, logistics means there are multiple use cases of Blockchain.
Among many prominent use cases of the technology, one is trade financing where Blockchain will be useful. So that will significantly reduce the complexity of the procession world, improve automation and optimisation of the processes. Hopefully, it will also reduce the cost of financing and that way it will probably open up the financing options to the untapped small and medium enterprises in India.
The other area which we are seeing a lot of interest is around supply chain financing. It is like invoice discounting and processes around that.
There is a lot of benefit in having those such type of processes mapped onto Blockchain because that will eliminate the need for any reconciliation of data across multiple entities such as banks, large manufacturers and their suppliers. It eliminates some of these redundant data being stored and disputes that arise.
A single record related to the information of all the transactions will help in telling about the errors in the processes. Are there any challenges in implementing this technology in India? In general, yes. But, in India specifically, we are unsure at this point. Generally, there are obviously challenges in early stages of a technology. There are definitely a lot of unknowns regards to the technology itself and how it can be adopted? There is business challenge in the sense that fundamentally, you try to get an ecosystem of participants together.
So, from a business standpoint, it needs multiple parties to cooperate and that always takes time. Connecting World, Empowering All The company started off with three employees in India when internet was in its infancy and there were limited web browsers and websites.
Juniper has been founded with one mission -Connect Everything and Empower Everyone. We are in the business of network infrastructure and we have three major business units namely routing, switching and security.
We work on three verticals that are service providing businesses, public-private sector and the enterprise sector includes BFSI. We recently celebrated 20 years in India and have the largest engineering presence in country. Juniper started off with three employees in India. At that point internet was in its infancy and there were very limited web browsers and websites.
Juniper saw a great opportunity in ensuring connectivity with everyone. We decided that internet protocol IP and ethnet a family of computer networking technologies commonly used in local area network are the two major technologies.
So we started creating products that will exponentially grow in all dimensions. We normally talk about three dimensions that are scale, subscriber base and services. Keeping this in mind we started building most scalable and service rich router to begin with. The most important things in a business are customer intimacy, understanding the real business need and translating that into technology solution.
There is a tendency of oversizing the network. A misconception is that the feature will be taken care of. So we have a network that can grow step-by-step, depending on the demand of scaleout. They emphasise more on the top players and make revenue out of that.
Telecom operators are of late gathering a lot of customer intelligence and then they provide personalised service. By , we will see two-time higher population of devices in India.
Nearly 25 to 30 million devices will come into picture. We are one of the leading vendors in the routing space and we are also one of the leading vendors in building cloud scale out data centres. We build devices that has two or three time greater capacity than competing vendors. How prepared are you for future challenges?
Technology is the need of the hour. We all have to move into automated systems. Skill set transformation for the existing engineers is a necessity, as this will develop their relevance in future. Do you have any important advice for your customers? Technology is revolutionary so you need to look beyond the traditional things. Look what your business vision is and start architecting the network to follow that part. Think beyond what is available.
Catalysts In Developing Technologies Red Hat remains the sole example of a pure play open source organisation, matching the revenue generated by even modest-sized proprietary alternatives, Hitesh Sahijwala, Director, Sales, Red Hat india Pvt. Red Hat is one of the largest open source organisations. If you look at our global reach, we are a 2 billion dollar company in terms of revenue, having presence across the world.
The philosophy of Red Hat is very clear. We are catalysts in the community of customers, partners and contributors to develop technologies which are State of the Art in the open source space. We harness the power of community to develop technology. I see Red Hat to be one of the leaders of the enterprises to open source segment. Are Indian banks willing to open source technologies against the proprietorship of softwares? Across the industry, the acceptance for Red Hat as an enterprise open source vendor or a partner is very high.
Today, Red Hat is not a part of one of the peripheral organisations partnering with the customers. Whether customers are running their core banking or running their own organisation systems, or running their CRM or customers like Bombay Stock Exchange, the volume of the trade has been phenomenally high.
The performance is times better than the proprietary system which they were getting earlier. Do you think there is still scope for open-end source systems? There is a lot of scope, if you look at the Banking and Finance sector. It is the largest adapter of this technology. Look at the government whether it is UID or my gov, they are all adopting open source. UID is running on RedHat platform. I do see very good adoption primarily because we are no more playing on a platform layer.
We are providing technology the middle way which helps integrating your disparate systems which were traditionally there. We also provide advance web services to integrate the advance channels to a customer environment. Speed to deploy is far more better and turnaround time is also far better. Because in proprietorship system you will be struggling within the four walls.
But here you can open into a community that will help you in fixing bugs. The performance of open source is times better than the proprietary system which they were getting earlier Do you require special kind of skill sets to handle these open source solutions?
If you look at the standardisation, process of development, process of maintenance are making seamless progress and this helps you in making things far more agile and adaptable.
Apart from Bombay Stock Exchange, what are the other clients you are catering to? We have lot of clients like Barclays Bank, Deutsche Bank who are using our open source technologies.
The main advantage is that you get far more enhancements in technologies, more expertise, better knowledge transfer that is why you are far more closer. Yes, definitely. And this is what we are covering up. The security layer is far more robust than the normal proprietary architecture.
All industries and segments are equally important for us. We devise focused strategies for them, considering the market dynamics of each one of them.
|Alcon official rebate form mailing address||663|
|Carefirst healthinsurance igbo translation||Centene corporation frmington|
|Deepak bhatia nuance transcription||Cigna adding newborn|
|Deepak bhatia nuance transcription||Nuance dragon education discount|
Write Review. A 4 Beds - 5 Baths - Villa. Punyisa Layan is a house and villa project located in Choeng Thale, Phuket and it is scheduled for completion in Jan It has 8 units and was developed by M2 Plus Co.
Available Units at Punyisa Layan. Get notified when new homes are being added to Punyisa Layan Create Alert. Price Rating : Great Price. Price Rating : Overpriced. Price Rating : Fair Price. Do you own a property at Punyisa Layan? Valuate Now How it works. Facilities and Floor Plans. Facilities Leisure Communal Garden Area. Safety 24H Security. Installment Milestone Payment. Finished foundation posts. Finished Floor,Built-ins,Windows,Doors. About the Developer - M2 Plus Co.
Frequently Asked Questions. What is the pet policy at Punyisa Layan? No pets are allowed at Punyisa Layan unless specifically permitted by the juristic office. Exceptions can be made for service or guide dogs for persons with disabilities. What and how far away is the nearest Beach from Punyisa Layan?
Layan Beach is the nearest Beach from Punyisa Layan and it is 2. How can foreigners own houses at Punyisa Layan? How does Leasehold work - is it safe? Some lease contracts do not allow lessee to have voting rights 3 - Is there a succession clause in the lease agreement that will allow inheritance of the lease? What is the process of purchasing a property at Punyisa Layan? Upon contract signing. Finished Structure.
Finished Walls. On handover. Reviews Know this project? Be the first to review. Neighborhood Overview. For Sale on FazWaz 1, properties. For Rent on FazWaz properties. To Blue Tree 6. To Phuket International Airport Location and Neighborhood. Surrounding area Punyisa Layan is located 1. Projects Nearby Punyisa Layan. His laboratory is associated with investigating mechanism-based chemopreventive and therapeutic effects of various medicinal plants, natural products, dietary and synthetic agents using various pre-clinical models of cancer.
Bhatia has extensively published original research papers and authoritative review articles, book chapters, and have presented over numerous papers at various national and international scientific meetings. Bhatia is the Editor of two journals and has acted as guest editor of numerous journal thematic issues. Bhatia is likewise serving as an editorial board member and ad-hoc reviewer of many reputed journals.
Transcriptional, post-transcriptional and epigenetic regulation of GADD45a. Bhatia is a recipient of: BJD Jr. Faculty research award. Follow Us.
Reservation Agreement. Reservation Deposit. Sales and Purchase Agreement. Payment Installments. Snag List. Punyisa Layan is located 1.
There are many other shops situated in the area of Punyisa Layan, the 3 closest being: - Lambretta Peugoet CF moto Phuket is m away about 2 mins. See the closest of them below: - Nan Lanna Food is 70 m away a walk of about 1 min. Phuket International Airport is located Request Details. Write Review. A 4 Beds - 5 Baths - Villa.
Punyisa Layan is a house and villa project located in Choeng Thale, Phuket and it is scheduled for completion in Jan It has 8 units and was developed by M2 Plus Co. Available Units at Punyisa Layan. Get notified when new homes are being added to Punyisa Layan Create Alert.
Price Rating : Great Price. Price Rating : Overpriced. Price Rating : Fair Price. Do you own a property at Punyisa Layan? Valuate Now How it works. Facilities and Floor Plans. Facilities Leisure Communal Garden Area. Safety 24H Security. Installment Milestone Payment. Finished foundation posts. Finished Floor,Built-ins,Windows,Doors. About the Developer - M2 Plus Co. Frequently Asked Questions.
What is the pet policy at Punyisa Layan? No pets are allowed at Punyisa Layan unless specifically permitted by the juristic office. Exceptions can be made for service or guide dogs for persons with disabilities. What and how far away is the nearest Beach from Punyisa Layan?
Layan Beach is the nearest Beach from Punyisa Layan and it is 2. How can foreigners own houses at Punyisa Layan? How does Leasehold work - is it safe? Some lease contracts do not allow lessee to have voting rights 3 - Is there a succession clause in the lease agreement that will allow inheritance of the lease? What is the process of purchasing a property at Punyisa Layan? Upon contract signing. Finished Structure. Finished Walls. On handover.
His laboratory is associated with investigating mechanism-based chemopreventive and therapeutic effects of various medicinal plants, natural products, dietary and synthetic agents using various pre-clinical models of cancer.
Bhatia has extensively published original research papers and authoritative review articles, book chapters, and have presented over numerous papers at various national and international scientific meetings. Bhatia is the Editor of two journals and has acted as guest editor of numerous journal thematic issues. Bhatia is likewise serving as an editorial board member and ad-hoc reviewer of many reputed journals. Transcriptional, post-transcriptional and epigenetic regulation of GADD45a.
Bhatia is a recipient of: BJD Jr. Faculty research award. Follow Us.