How to Do Things With Words

We propose a unified neural network architecture and learning algorithm that can be applied to various natural language processing tasks including part-of-speech tagging, chunking, named entity recognition, and semantic role labeling. This versatility is achieved by trying to avoid task-specific engineering and therefore disregarding a lot of prior knowledge. Instead of exploiting man-made input features carefully optimized for each task, our system learns internal representations on the basis of vast amounts of mostly unlabeled training data. This work is then used as a basis for building a freely available tagging system with good performance and minimal computational requirements.

/pdf/natural-language-processing-almost-from-scratch-osj5vem4f3.pdf

Natural Language Processing (Almost) from Scratch

This paper introduces a new corpus of read English speech, suitable for training and evaluating speech recognition systems. The LibriSpeech corpus is derived from audiobooks that are part of the LibriVox project, and contains 1000 hours of speech sampled at 16 kHz. We have made the corpus freely available for download, along with separately prepared language-model training data and pre-built language models. We show that acoustic models trained on LibriSpeech give lower error rate on the Wall Street Journal (WSJ) test sets than models trained on WSJ itself. We are also releasing Kaldi scripts that make it easy to build these systems.

/pdf/librispeech-an-asr-corpus-based-on-public-domain-audio-books-gslbn35u4y.pdf

Librispeech: An ASR corpus based on public domain audio books

Strategies of Discourse Comprehension

GROUNDING It takes two people working together to play a duet, shake hands, play chess, waltz, teach, or make love. To succeed, the two of them have to coordinate both the content and process of what they are doing. Alan and Barbara, on the piano, must come to play the same Mozart duet. This is coordination of content. They must also synchronize their entrances and exits, coordinate how loudly to play forte and pianissimo, and otherwise adjust to each other's tempo and dynamics. This is coordination of process. They cannot even begin to coordinate on content without assuming a vast amount of shared information or common ground-that is, mutual knowledge, mutual beliefs, and mutual assumptions And to coordinate on process, they need to update their common ground moment by moment. All collective actions are built on common ground and its accumulation. We thank many colleagues for discussion of the issues we take up here.

/pdf/grounding-in-communication-2do08wtssx.pdf

Grounding in communication

This paper concerns relationships among focus of attention, choice of referring expression, and perceived coherence of utterances within a discourse segment. It presents a framework and initial theory of centering intended to model the local component of attentional state. The paper examines interactions between local coherence and choice of referring expressions; it argues that differences in coherence correspond in part to the inference demands made by different types of referring expressions, given a particular attentional state. It demonstrates that the attentional state properties modeled by centering can account for these differences.

/pdf/centering-a-framework-for-modeling-the-local-coherence-of-hvb4n7dktx.pdf

Centering: a framework for modeling the local coherence of discourse

We present the second version of the Penn Discourse Treebank, PDTB-2.0, describing its lexically-grounded annotations of discourse relations and their two abstract object arguments over the 1 million word Wall Street Journal corpus. We describe all aspects of the annotation, including (a) the argument structure of discourse relations, (b) the sense annotation of the relations, and (c) the attribution of discourse relations and each of their arguments. We list the differences between PDTB-1.0 and PDTB-2.0. We present representative statistics for several aspects of the annotation in the corpus.

/pdf/the-penn-discourse-treebank-2-0-lkwywektwx.pdf

The Penn Discourse TreeBank 2.0.

Tree adjunct grammars

In this paper, we will describe a tree generating system called tree-adjoining grammar (TAG) and state some of the recent results about TAGs. The work on TAGs is motivated by linguistic considerations. However, a number of formal results have been established for TAGs, which we believe, would be of interest to researchers in formal languages and automata, including those interested in tree grammars and tree automata.

Aravind K. Joshi

Papers

Centering: a framework for modeling the local coherence of discourse

The Penn Discourse TreeBank 2.0.

Tree adjunct grammars

Tree-adjoining grammars

Elements of Discourse Understanding