Search (49 results, page 2 of 3)

0.0023042266 = product of:
  0.01382536 = sum of:
    0.01382536 = weight(_text_:in in 2804) [ClassicSimilarity], result of:
      0.01382536 = score(doc=2804,freq=30.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.23282567 = fieldWeight in 2804, product of:
          5.477226 = tf(freq=30.0), with freq of:
            30.0 = termFreq=30.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.03125 = fieldNorm(doc=2804)
  0.16666667 = coord(1/6)

Abstract

The ACL Anthology is a digital archive of conference and journal papers in natural language processing and computational linguistics. Its primary purpose is to serve as a reference repository of research results, but we believe that it can also be an object of study and a platform for research in its own right. We describe an enriched and standardized reference corpus derived from the ACL Anthology that can be used for research in scholarly document processing. This corpus, which we call the ACL Anthology Reference Corpus (ACL ARC), brings together the recent activities of a number of research groups around the world. Our goal is to make the corpus widely available, and to encourage other researchers to use it as a standard testbed for experiments in both bibliographic and bibliometric research.

Content

Vgl. auch: Automatic Term Recognition (ATR) is a research task that deals with the identification of domain-specific terms. Terms, in simple words, are textual realization of significant concepts in an expertise domain. Additionally, domain-specific terms may be classified into a number of categories, in which each category represents a significant concept. A term classification task is often defined on top of an ATR procedure to perform such categorization. For instance, in the biomedical domain, terms can be classified as drugs, proteins, and genes. This is a reference dataset for terminology extraction and classification research in computational linguistics. It is a set of manually annotated terms in English language that are extracted from the ACL Anthology Reference Corpus (ACL ARC). The ACL ARC is a canonicalised and frozen subset of scientific publications in the domain of Human Language Technologies (HLT). It consists of 10,921 articles from 1965 to 2006. The dataset, called ACL RD-TEC, is comprised of more than 69,000 candidate terms that are manually annotated as valid and invalid terms. Furthermore, valid terms are classified as technology and non-technology terms. Technology terms refer to a method, process, or in general a technological concept in the domain of HLT, e.g. machine translation, word sense disambiguation, and language modelling. On the other hand, non-technology terms refer to important concepts other than technological; examples of such terms in the domain of HLT are multilingual lexicon, corpora, word sense, and language model. The dataset is created to serve as a gold standard for the comparison of the algorithms of term recognition and classification. [http://catalog.elra.info/product_info.php?products_id=1236].

0.0021859813 = product of:
  0.013115887 = sum of:
    0.013115887 = weight(_text_:in in 330) [ClassicSimilarity], result of:
      0.013115887 = score(doc=330,freq=12.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.22087781 = fieldWeight in 330, product of:
          3.4641016 = tf(freq=12.0), with freq of:
            12.0 = termFreq=12.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=330)
  0.16666667 = coord(1/6)

Abstract

The complexity of human interactions with social and natural phenomena is mirrored in the way we describe our experiences through natural language. In order to retain and convey such a high dimensional information, the statistical properties of our linguistic output has to be highly correlated in time. An example are the robust observations, still largely not understood, of correlations on arbitrary long scales in literary texts. In this paper we explain how long-range correlations flow from highly structured linguistic levels down to the building blocks of a text (words, letters, etc..). By combining calculations and data analysis we show that correlations take form of a bursty sequence of events once we approach the semantically relevant topics of the text. The mechanisms we identify are fairly general and can be equally applied to other hierarchical settings.

0.0021034614 = product of:
  0.012620768 = sum of:
    0.012620768 = weight(_text_:in in 925) [ClassicSimilarity], result of:
      0.012620768 = score(doc=925,freq=4.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.21253976 = fieldWeight in 925, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.078125 = fieldNorm(doc=925)
  0.16666667 = coord(1/6)

Abstract

Mit KI lassen sich in kurzer Zeit Texte erzeugen. Ob Studierende die Programme einsetzen dürfen, ist nicht einheitlich geregelt. Oft hängt die Entscheidung am Lehrpersonal, wie eine BR-Umfrage zeigt.

0.0020823204 = product of:
  0.012493922 = sum of:
    0.012493922 = weight(_text_:in in 4121) [ClassicSimilarity], result of:
      0.012493922 = score(doc=4121,freq=8.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.21040362 = fieldWeight in 4121, product of:
          2.828427 = tf(freq=8.0), with freq of:
            8.0 = termFreq=8.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0546875 = fieldNorm(doc=4121)
  0.16666667 = coord(1/6)

Abstract

Intuition suggests that one way to enhance the information retrieval process would be the use of phrases to characterize the contents of text. A number of researchers, however, have noted that phrases alone do not improve retrieval effectiveness. In this paper we briefly review the use of phrases in information retrieval and then suggest extensions to this paradigm using semantic information. We claim that semantic processing, which can be viewed as expressing relations between the concepts represented by phrases, will in fact enhance retrieval effectiveness. The availability of the UMLS® domain model, which we exploit extensively, significantly contributes to the feasibility of this processing.

0.0020823204 = product of:
  0.012493922 = sum of:
    0.012493922 = weight(_text_:in in 668) [ClassicSimilarity], result of:
      0.012493922 = score(doc=668,freq=8.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.21040362 = fieldWeight in 668, product of:
          2.828427 = tf(freq=8.0), with freq of:
            8.0 = termFreq=8.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0546875 = fieldNorm(doc=668)
  0.16666667 = coord(1/6)

Abstract

We investigate different systems for extracting mathematical entities from English texts in the mathematical field of category theory as a first step for constructing a mathematical knowledge graph. We consider four different term extractors and compare their results. This small experiment showcases some of the issues with the construction and evaluation of terms extracted from noisy domain text. We also make available two open corpora in research mathematics, in particular in category theory: a small corpus of 755 abstracts from the journal TAC (3188 sentences), and a larger corpus from the nLab community wiki (15,000 sentences).

0.0019955188 = product of:
  0.011973113 = sum of:
    0.011973113 = weight(_text_:in in 2803) [ClassicSimilarity], result of:
      0.011973113 = score(doc=2803,freq=10.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.20163295 = fieldWeight in 2803, product of:
          3.1622777 = tf(freq=10.0), with freq of:
            10.0 = termFreq=10.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=2803)
  0.16666667 = coord(1/6)

Abstract

This paper introduces ACL RD-TEC: a dataset for evaluating the extraction and classification of terms from literature in the domain of computational linguistics. The dataset is derived from the Association for Computational Linguistics anthology reference corpus (ACL ARC). In its first release, the ACL RD-TEC consists of automatically segmented, part-of-speech-tagged ACL ARC documents, three lists of candidate terms, and more than 82,000 manually annotated terms. The annotated terms are marked as either valid or invalid, and valid terms are further classified as technology and non-technology terms. Technology terms signify methods, algorithms, and solutions in computational linguistics. The paper describes the dataset and reports the relevant statistics. We hope the step described in this paper encourages a collaborative effort towards building a full-fledged annotated corpus from the computational linguistics literature.

0.0018033426 = product of:
  0.010820055 = sum of:
    0.010820055 = weight(_text_:in in 667) [ClassicSimilarity], result of:
      0.010820055 = score(doc=667,freq=6.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.1822149 = fieldWeight in 667, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0546875 = fieldNorm(doc=667)
  0.16666667 = coord(1/6)

Abstract

We present an overview of the NTCIR Math Tasks organized during NTCIR-10, 11, and 12. These tasks are primarily dedicated to techniques for searching mathematical content with formula expressions. In this chapter, we first summarize the task design and introduce test collections generated in the tasks. We also describe the features and main challenges of mathematical information retrieval systems and discuss future perspectives in the field.

0.0017848461 = product of:
  0.010709076 = sum of:
    0.010709076 = weight(_text_:in in 2697) [ClassicSimilarity], result of:
      0.010709076 = score(doc=2697,freq=8.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.18034597 = fieldWeight in 2697, product of:
          2.828427 = tf(freq=8.0), with freq of:
            8.0 = termFreq=8.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=2697)
  0.16666667 = coord(1/6)

Abstract

Text mining and natural language processing are fast growing areas of research, with numerous applications in business, science and creative industries. This paper presents TextFlows, a web-based text mining and natural language processing platform supporting workflow construction, sharing and execution. The platform enables visual construction of text mining workflows through a web browser, and the execution of the constructed workflows on a processing cloud. This makes TextFlows an adaptable infrastructure for the construction and sharing of text processing workflows, which can be reused in various applications. The paper presents the implemented text mining and language processing modules, and describes some precomposed workflows. Their features are demonstrated on three use cases: comparison of document classifiers and of different part-of-speech taggers on a text categorization problem, and outlier detection in document corpora.

Source

Science of computer programming. In Press, 2016

0.001682769 = product of:
  0.010096614 = sum of:
    0.010096614 = weight(_text_:in in 1161) [ClassicSimilarity], result of:
      0.010096614 = score(doc=1161,freq=4.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.17003182 = fieldWeight in 1161, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0625 = fieldNorm(doc=1161)
  0.16666667 = coord(1/6)

Abstract

We present a novel unsupervised approach to detecting the compositionality of multi-word expressions. We compute the compositionality of a phrase through substituting the constituent words with their "neighbours" in a semantic vector space and averaging over the distance between the original phrase and the substituted neighbour phrases. Several methods of obtaining neighbours are presented. The results are compared to existing supervised results and achieve state-of-the-art performance on a verb-object dataset of human compositionality ratings.

0.0015457221 = product of:
  0.009274333 = sum of:
    0.009274333 = weight(_text_:in in 1061) [ClassicSimilarity], result of:
      0.009274333 = score(doc=1061,freq=6.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.1561842 = fieldWeight in 1061, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=1061)
  0.16666667 = coord(1/6)

Abstract

The object of Information Retrieval is to retrieve all relevant documents for a user query and only those relevant documents. Much research has focused on achieving this objective with little regard for storage overhead or performance. In the paper we evaluate the use of Part of Speech Tagging to improve, the index storage overhead and general speed of the system with only a minimal reduction to precision recall measurements. We tagged 500Mbs of the Los Angeles Times 1990 and 1989 document collection provided by TREC for parts of speech. We then experimented to find the most relevant part of speech to index. We show that 90% of precision recall is achieved with 40% of the document collections terms. We also show that this is a improvement in overhead with only a 1% reduction in precision recall.

0.0015457221 = product of:
  0.009274333 = sum of:
    0.009274333 = weight(_text_:in in 22) [ClassicSimilarity], result of:
      0.009274333 = score(doc=22,freq=6.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.1561842 = fieldWeight in 22, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=22)
  0.16666667 = coord(1/6)

Abstract

DBS uses a strictly time-linear derivation order. Therefore the basic computational complexity degree of DBS is linear time. The only way to increase DBS complexity above linear is repeating ambiguity. In natural language, however, repeating ambiguity is prevented by grammatical disambiguation. A classic example of a grammatical ambiguity is the 'garden path' sentence The horse raced by the barn fell. The continuation horse+raced introduces an ambiguity between horse which raced and horse which was raced, leading to two parallel derivation strands up to The horse raced by the barn. Depending on whether the continuation is interpunctuation or a verb, they are grammatically disambiguated, resulting in unambiguous output. A repeated ambiguity occurs in The man who loves the woman who feeds Lucy who Peter loves., with who serving as subject or as object. These readings are grammatically disambiguated by continuing after who with a verb or a noun.

0.0014873719 = product of:
  0.008924231 = sum of:
    0.008924231 = weight(_text_:in in 864) [ClassicSimilarity], result of:
      0.008924231 = score(doc=864,freq=2.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.15028831 = fieldWeight in 864, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.078125 = fieldNorm(doc=864)
  0.16666667 = coord(1/6)

Abstract

Dialoge sind in. Dieser Trend hat mit dem Chatbot LaMDA auch Google erreicht. Die Zukunft der Suchanfrage ist das Gespräch, das Gespräch mit dem Objekt der Neugier. Ob das wirklich eine gute Idee ist, fragt sich Medienphilosoph Roberto Simanowski.

0.0014873719 = product of:
  0.008924231 = sum of:
    0.008924231 = weight(_text_:in in 927) [ClassicSimilarity], result of:
      0.008924231 = score(doc=927,freq=2.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.15028831 = fieldWeight in 927, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.078125 = fieldNorm(doc=927)
  0.16666667 = coord(1/6)

Abstract

Bis vor kurzem völlig undenkbar: Eine mit ChatGPT vergleichbare Sprach-KI läuft lokal (!) auf Standard-Rechnern. c't 3003 hat's ausprobiert. Eine Sprach-KI in 4,2 Gigabyte? Wie soll das denn gehen?

0.0014724231 = product of:
  0.008834538 = sum of:
    0.008834538 = weight(_text_:in in 3922) [ClassicSimilarity], result of:
      0.008834538 = score(doc=3922,freq=4.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.14877784 = fieldWeight in 3922, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0546875 = fieldNorm(doc=3922)
  0.16666667 = coord(1/6)

Abstract

WordHoard defines a multiword unit as a special type of collocate in which the component words comprise a meaningful phrase. For example, "Knight of the Round Table" is a meaningful multiword unit or phrase. WordHoard uses the notion of a pseudo-bigram to generalize the computation of bigram (two word) statistical measures to phrases (n-grams) longer than two words, and to allow comparisons of these measures for phrases with different word counts. WordHoard applies the localmaxs algorithm of Silva et al. to the pseudo-bigrams to identify potential compositional phrases that "stand out" in a text. WordHoard can also filter two and three word phrases using the word class filters suggested by Justeson and Katz.

0.0014724231 = product of:
  0.008834538 = sum of:
    0.008834538 = weight(_text_:in in 156) [ClassicSimilarity], result of:
      0.008834538 = score(doc=156,freq=4.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.14877784 = fieldWeight in 156, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0546875 = fieldNorm(doc=156)
  0.16666667 = coord(1/6)

Abstract

In this work, we present the results of research for evaluating a methodology for extracting mathematical terms for precalculus using the techniques for semantically-oriented statistical search. We use the corpus-based approach and the combination of different statistically-based techniques for extracting keywords, collocations and co-occurrences incorporated in the Sketch Engine software. We evaluate the collocations candidate terms for the basic concept function(s) and approve the related methodology by precalculus domain conceptual terms definitions. Finally, we offer a conceptual terms hierarchical representation and discuss the results with respect to their possible applications.

0.0014724231 = product of:
  0.008834538 = sum of:
    0.008834538 = weight(_text_:in in 1123) [ClassicSimilarity], result of:
      0.008834538 = score(doc=1123,freq=4.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.14877784 = fieldWeight in 1123, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0546875 = fieldNorm(doc=1123)
  0.16666667 = coord(1/6)

Abstract

WordHoard defines a multiword unit as a special type of collocate in which the component words comprise a meaningful phrase. For example, "Knight of the Round Table" is a meaningful multiword unit or phrase. WordHoard uses the notion of a pseudo-bigram to generalize the computation of bigram (two word) statistical measures to phrases (n-grams) longer than two words, and to allow comparisons of these measures for phrases with different word counts. WordHoard applies the localmaxs algorithm of Silva et al. to the pseudo-bigrams to identify potential compositional phrases that "stand out" in a text. WordHoard can also filter two and three word phrases using the word class filters suggested by Justeson and Katz.

0.0014724231 = product of:
  0.008834538 = sum of:
    0.008834538 = weight(_text_:in in 1150) [ClassicSimilarity], result of:
      0.008834538 = score(doc=1150,freq=4.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.14877784 = fieldWeight in 1150, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0546875 = fieldNorm(doc=1150)
  0.16666667 = coord(1/6)

Abstract

Voice search is becoming a popular mode for interacting with search engines. As a result, research has gone into building better voice transcription engines, interfaces, and search engines that better handle inherent verbosity of queries. However, when one considers its use by non- native speakers of English, another aspect that becomes important is the formulation of the query by users. In this paper, we present the results of a preliminary study that we conducted with non-native English speakers who formulate queries for given retrieval tasks. Our results show that the current search engines are sensitive in their rankings to the query formulation, and thus highlights the need for developing more robust ranking methods.

0.0014724231 = product of:
  0.008834538 = sum of:
    0.008834538 = weight(_text_:in in 773) [ClassicSimilarity], result of:
      0.008834538 = score(doc=773,freq=4.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.14877784 = fieldWeight in 773, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0546875 = fieldNorm(doc=773)
  0.16666667 = coord(1/6)

Abstract

We show that generating English Wikipedia articles can be approached as a multi-document summarization of source documents. We use extractive summarization to coarsely identify salient information and a neural abstractive model to generate the article. For the abstractive model, we introduce a decoder-only architecture that can scalably attend to very long sequences, much longer than typical encoder- decoder architectures used in sequence transduction. We show that this model can generate fluent, coherent multi-sentence paragraphs and even whole Wikipedia articles. When given reference documents, we show it can extract relevant factual information as reflected in perplexity, ROUGE scores and human evaluations.

0.0012881019 = product of:
  0.007728611 = sum of:
    0.007728611 = weight(_text_:in in 2861) [ClassicSimilarity], result of:
      0.007728611 = score(doc=2861,freq=6.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.1301535 = fieldWeight in 2861, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=2861)
  0.16666667 = coord(1/6)

Abstract

Today's conventional search engines hardly do provide the essential content relevant to the user's search query. This is because the context and semantics of the request made by the user is not analyzed to the full extent. So here the need for a semantic web search arises. SWS is upcoming in the area of web search which combines Natural Language Processing and Artificial Intelligence. The objective of the work done here is to design, develop and implement a semantic search engine- SIEU(Semantic Information Extraction in University Domain) confined to the university domain. SIEU uses ontology as a knowledge base for the information retrieval process. It is not just a mere keyword search. It is one layer above what Google or any other search engines retrieve by analyzing just the keywords. Here the query is analyzed both syntactically and semantically. The developed system retrieves the web results more relevant to the user query through keyword expansion. The results obtained here will be accurate enough to satisfy the request made by the user. The level of accuracy will be enhanced since the query is analyzed semantically. The system will be of great use to the developers and researchers who work on web. The Google results are re-ranked and optimized for providing the relevant links. For ranking an algorithm has been applied which fetches more apt results for the user query.

0.0012620769 = product of:
  0.0075724614 = sum of:
    0.0075724614 = weight(_text_:in in 869) [ClassicSimilarity], result of:
      0.0075724614 = score(doc=869,freq=4.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.12752387 = fieldWeight in 869, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=869)
  0.16666667 = coord(1/6)

Abstract

Generative Pre-trained Transformer (GPT) models by OpenAI have taken natural language processing (NLP) community by storm by introducing very powerful language models. These models can perform various NLP tasks like question answering, textual entailment, text summarisation etc. without any supervised training. These language models need very few to no examples to understand the tasks and perform equivalent or even better than the state-of-the-art models trained in supervised fashion. In this article we will cover the journey of these models and understand how they have evolved over a period of 2 years. 1. Discussion of GPT-1 paper (Improving Language Understanding by Generative Pre-training). 2. Discussion of GPT-2 paper (Language Models are unsupervised multitask learners) and its subsequent improvements over GPT-1. 3. Discussion of GPT-3 paper (Language models are few shot learners) and the improvements which have made it one of the most powerful models NLP has seen till date. This article assumes familiarity with the basics of NLP terminologies and transformer architecture.