Dipanjan Das

A Well-Composed Text is Half Done! Composition Sampling for Diverse Conditional Generation

2022 · arXiv (Cornell University)

We propose Composition Sampling, a simple but effective method to generate diverse outputs for conditional generation of higher quality compared to previous stochastic decoding strategies. It builds on recently proposed plan-based neural generation models (Narayan …

Part-of-Speech Tagging for Twitter: Annotation, Features, and Experiments

2018 · Figshare

We address the problem of part-of-speech tagging for English data from the popular microblogging service Twitter. We develop a tagset, annotate data, develop features, and report tagging results nearing 90% accuracy. The data and tools …

Transforming Dependency Structures to Logical Forms for Semantic Parsing

2016 · Transactions of the Association for Computational Linguistics

The strongly typed syntax of grammar formalisms such as CCG, TAG, LFG and HPSG offers a synchronous framework for deriving syntactic structures and semantic logical forms. In contrast—partly due to the lack of a strong …

A Decomposable Attention Model for Natural Language Inference

2016 · arXiv (Cornell University)

We propose a simple neural architecture for natural language inference.Our approach uses attention to decompose the problem into subproblems that can be solved separately, thus making it trivially parallelizable.On the Stanford Natural Language Inference (SNLI) …

Learning Recurrent Span Representations for Extractive Question Answering

2016 · arXiv (Cornell University)

The reading comprehension task, that asks questions about a given evidence document, is a central problem in natural language understanding. Recent formulations of this task have typically focused on answer selection from a set of …

Neural Paraphrase Identification of Questions with Noisy Pretraining

2017

We present a solution to the problem of paraphrase identification of questions. We focus on a recent dataset of question pairs annotated with binary paraphrase labels and show that a variant of the decomposable attention …

BERT Rediscovers the Classical NLP Pipeline

2019

Pre-trained text encoders have rapidly advanced the state of the art on many NLP tasks. We focus on one such model, BERT, and aim to quantify where linguistic information is captured within the network. We …

BLEURT: Learning Robust Metrics for Text Generation

2020

Text generation has made significant advances in the last few years. Yet, evaluation metrics have lagged behind, as the most popular choices (e.g., BLEU and ROUGE) may correlate poorly with human judgments. We propose BLEURT, …

Increasing Faithfulness in Knowledge-Grounded Dialogue with Controllable Features

2021

Hannah Rashkin, David Reitter, Gaurav Singh Tomar, Dipanjan Das. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). …

The GEM Benchmark: Natural Language Generation, its Evaluation and Metrics

2021

Sebastian Gehrmann, Tosin Adewumi, Karmanya Aggarwal, Pawan Sasanka Ammanamanchi, Anuoluwapo Aremu, Antoine Bosselut, Khyathi Raghavi Chandu, Miruna-Adriana Clinciu, Dipanjan Das, Kaustubh Dhole, Wanyu Du, Esin Durmus, Ondřej Dušek, Chris Chinenye Emezue, Varun Gangal, Cristina Garbacea, …

What do you learn from context? Probing for sentence structure in\n contextualized word representations

2019 · arXiv (Cornell University)

Contextualized representation models such as ELMo (Peters et al., 2018a) and\nBERT (Devlin et al., 2018) have recently achieved state-of-the-art results on a\ndiverse array of downstream NLP tasks. Building on recent token-level probing\nwork, we introduce a …