Mitchell Marcus

Bio: Mitchell Marcus is an academic researcher from University of Pennsylvania. The author has contributed to research in topics: Parsing & Natural language. The author has an hindex of 33, co-authored 71 publications receiving 17799 citations. Previous affiliations of Mitchell Marcus include Massachusetts Institute of Technology & Cornell University.

Building a large annotated corpus of English: the penn treebank

Abstract: : As a result of this grant, the researchers have now published oil CDROM a corpus of over 4 million words of running text annotated with part-of- speech (POS) tags, with over 3 million words of that material assigned skeletal grammatical structure This material now includes a fully hand-parsed version of the classic Brown corpus About one half of the papers at the ACL Workshop on Using Large Text Corpora this past summer were based on the materials generated by this grant

System for generation of user profiles for a system for customized electronic identification of desirable objects

Abstract: This invention relates to customized electronic identification of desirable objects, such as news articles, in an electronic media environment, and in particular to a system that automatically constructs both a "target profile" for each target object in the electronic media based, for example, on the frequency with which each word appears in an article relative to its overall frequency of use in all articles, as well as a "target profile interest summary" for each user, which target profile interest summary describes the user's interest level in various types of target objects. The system then evaluates the target profiles against the users' target profile interest summaries to generate a user-customized rank ordered listing of target objects most likely to be of interest to each user so that the user can select from among these potentially relevant target objects, which were automatically selected by this system from the plethora of target objects that are profiled on the electronic media. Users' target profile interest summaries can be used to efficiently organize the distribution of information in a large scale system consisting of many users interconnected by means of a communication network. Additionally, a cryptographically-based pseudonym proxy server is provided to ensure the privacy of a user's target profile interest summary, by giving the user control over the ability of third parties to access this summary and to identify or contact the user.

Text Chunking Using Transformation-Based Learning

Abstract: Transformation-based learning, a technique introduced by Eric Brill (1993b), has been shown to do part-of-speech tagging with fairly high accuracy. This same method can be applied at a higher level of textual interpretation for locating chunks in the tagged text, including non-recursive “baseNP” chunks. For this purpose, it is convenient to view chunking as a tagging problem by encoding the chunk structure in new tags attached to each word. In automatic tests using Treebank-derived data, this technique achieved recall and precision rates of roughly 93% for baseNP chunks (trained on 950K words) and 88% for somewhat more complex chunks that partition the sentence (trained on 200K words). Working in this new application and with larger template and training sets has also required some interesting adaptations to the transformation-based learning approach.

Text Chunking using Transformation-Based Learning

Abstract: Eric Brill introduced transformation-based learning and showed that it can do part-of-speech tagging with fairly high accuracy. The same method can be applied at a higher level of textual interpretation for locating chunks in the tagged text, including non-recursive ``baseNP'' chunks. For this purpose, it is convenient to view chunking as a tagging problem by encoding the chunk structure in new tags attached to each word. In automatic tests using Treebank-derived data, this technique achieved recall and precision rates of roughly 92% for baseNP chunks and 88% for somewhat more complex chunks that partition the sentence. Some interesting adaptations to the transformation-based learning approach are also suggested by this application.

OntoNotes: The 90% Solution

Abstract: We describe the OntoNotes methodology and its result, a large multilingual richly-annotated corpus constructed at 90% interannotator agreement. An initial portion (300K words of English newswire and 250K words of Chinese newswire) will be made available to the community during 2007.

Language Models are Few-Shot Learners

Abstract: Recent work has demonstrated substantial gains on many NLP tasks and benchmarks by pre-training on a large corpus of text followed by fine-tuning on a specific task. While typically task-agnostic in architecture, this method still requires task-specific fine-tuning datasets of thousands or tens of thousands of examples. By contrast, humans can generally perform a new language task from only a few examples or from simple instructions - something which current NLP systems still largely struggle to do. Here we show that scaling up language models greatly improves task-agnostic, few-shot performance, sometimes even reaching competitiveness with prior state-of-the-art fine-tuning approaches. Specifically, we train GPT-3, an autoregressive language model with 175 billion parameters, 10x more than any previous non-sparse language model, and test its performance in the few-shot setting. For all tasks, GPT-3 is applied without any gradient updates or fine-tuning, with tasks and few-shot demonstrations specified purely via text interaction with the model. GPT-3 achieves strong performance on many NLP datasets, including translation, question-answering, and cloze tasks, as well as several tasks that require on-the-fly reasoning or domain adaptation, such as unscrambling words, using a novel word in a sentence, or performing 3-digit arithmetic. At the same time, we also identify some datasets where GPT-3's few-shot learning still struggles, as well as some datasets where GPT-3 faces methodological issues related to training on large web corpora. Finally, we find that GPT-3 can generate samples of news articles which human evaluators have difficulty distinguishing from articles written by humans. We discuss broader societal impacts of this finding and of GPT-3 in general.

Foundations of Statistical Natural Language Processing

Abstract: Statistical approaches to processing natural language text have become dominant in recent years This foundational text is the first comprehensive introduction to statistical natural language processing (NLP) to appear The book contains all the theory and algorithms needed for building NLP tools It provides broad but rigorous coverage of mathematical and linguistic foundations, as well as detailed discussion of statistical methods, allowing students and researchers to construct their own implementations The book covers collocation finding, word sense disambiguation, probabilistic parsing, information retrieval, and other applications

Building a large annotated corpus of English: the penn treebank

Abstract: : As a result of this grant, the researchers have now published oil CDROM a corpus of over 4 million words of running text annotated with part-of- speech (POS) tags, with over 3 million words of that material assigned skeletal grammatical structure This material now includes a fully hand-parsed version of the classic Brown corpus About one half of the papers at the ACL Workshop on Using Large Text Corpora this past summer were based on the materials generated by this grant

Deep contextualized word representations

Abstract: We introduce a new type of deep contextualized word representation that models both (1) complex characteristics of word use (e.g., syntax and semantics), and (2) how these uses vary across linguistic contexts (i.e., to model polysemy). Our word vectors are learned functions of the internal states of a deep bidirectional language model (biLM), which is pre-trained on a large text corpus. We show that these representations can be easily added to existing models and significantly improve the state of the art across six challenging NLP problems, including question answering, textual entailment and sentiment analysis. We also present an analysis showing that exposing the deep internals of the pre-trained network is crucial, allowing downstream models to mix different types of semi-supervision signals.

Attention Is All You Need

Abstract: The dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.8 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data.