1 code implementation • LREC 2022 • Sayontan Ghosh, Amanpreet Singh, Alex Merenstein, Wei Su, Scott A. Smolka, Erez Zadok, Niranjan Balasubramanian
Evaluations show that even when using a state-of-the-art language model, there is significant room for improvement, with the best models achieving an F1 score of only 60. 5 and 33. 3 in the named-entity-recognition and dependency-link-prediction sub-tasks, respectively.
1 code implementation • EMNLP (ACL) 2021 • Yash Kumar Lal, Reetu Singh, Harsh Trivedi, Qingqing Cao, Aruna Balasubramanian, Niranjan Balasubramanian
IrEne is an energy prediction system that accurately predicts the interpretable inference energy consumption of a wide range of Transformer-based NLP models.
1 code implementation • 14 Feb 2023 • Mahnaz Koupaee, Greg Durrett, Nathanael Chambers, Niranjan Balasubramanian
Event scenarios are often complex and involve multiple event sequences connected through different entity participants.
1 code implementation • 20 Dec 2022 • Harsh Trivedi, Niranjan Balasubramanian, Tushar Khot, Ashish Sabharwal
This is insufficient, however, when the necessary knowledge is not available or up-to-date within a model's parameters.
1 code implementation • 5 Dec 2022 • Sai Vallurupalli, Sayontan Ghosh, Katrin Erk, Niranjan Balasubramanian, Francis Ferraro
Knowledge about outcomes is critical for complex event understanding but is hard to acquire.
1 code implementation • 26 Oct 2022 • Mohaddeseh Bastan, Mihai Surdeanu, Niranjan Balasubramanian
We introduce a novel semi-supervised procedure that bootstraps an NLI dataset from existing biomedical dataset that pairs mechanisms with experimental evidence in abstracts.
Ranked #1 on
Natural Language Inference
on BioNLI
no code implementations • 12 Oct 2022 • Sayontan Ghosh, Tanvi Aggarwal, Minh Hoai, Niranjan Balasubramanian
Anticipating future actions in a video is useful for many autonomous and assistive technologies.
no code implementations • 31 Aug 2022 • Marcos Treviso, Ji-Ung Lee, Tianchu Ji, Betty van Aken, Qingqing Cao, Manuel R. Ciosici, Michael Hassid, Kenneth Heafield, Sara Hooker, Colin Raffel, Pedro H. Martins, André F. T. Martins, Jessica Zosa Forde, Peter Milder, Edwin Simpson, Noam Slonim, Jesse Dodge, Emma Strubell, Niranjan Balasubramanian, Leon Derczynski, Iryna Gurevych, Roy Schwartz
Recent work in natural language processing (NLP) has yielded appealing results from scaling model parameters and training data; however, using only scale to improve performance means that resource consumption also grows.
no code implementations • 31 Jul 2022 • Sayontan Ghosh, Mahnaz Koupaee, Isabella Chen, Francis Ferraro, Nathanael Chambers, Niranjan Balasubramanian
Often, these participant states are not explicitly mentioned in the narrative, left to be filled in via common-sense or inference.
1 code implementation • 25 May 2022 • Harsh Trivedi, Niranjan Balasubramanian, Tushar Khot, Ashish Sabharwal
We show how to use question decompositions to teach language models these broad reasoning skills in a robust fashion.
no code implementations • 24 May 2022 • Xueying Bai, Jinghuan Shang, Yifan Sun, Niranjan Balasubramanian
Continual learning (CL) aims to learn a sequence of tasks over time, with data distributions shifting from one task to another.
2 code implementations • ACL ARR November 2021 • Mohaddeseh Bastan, Nishant Shankar, Mihai Surdeanu, Niranjan Balasubramanian
We leverage this structure and create a summarization task, where the input is a collection of sentences and the main entities in an abstract, and the output includes the relationship and a sentence that summarizes the mechanism.
1 code implementation • Findings (ACL) 2022 • Nikita Soni, Matthew Matero, Niranjan Balasubramanian, H. Andrew Schwartz
Natural language is generated by people, yet traditional language modeling views words or documents as if generated independently.
1 code implementation • Findings (EMNLP) 2021 • Matthew Matero, Nikita Soni, Niranjan Balasubramanian, H. Andrew Schwartz
Much of natural language processing is focused on leveraging large capacity language models, typically trained over single messages with a task of predicting one or more tokens.
1 code implementation • EMNLP 2021 • Naoya Inoue, Harsh Trivedi, Steven Sinha, Niranjan Balasubramanian, Kentaro Inui
Instead, we advocate for an abstractive approach, where we propose to generate a question-focused, abstractive summary of input paragraphs and then feed it to an RC system.
1 code implementation • 2 Aug 2021 • Harsh Trivedi, Niranjan Balasubramanian, Tushar Khot, Ashish Sabharwal
Multihop reasoning remains an elusive goal as existing multihop benchmarks are known to be largely solvable via shortcuts.
no code implementations • ACL 2021 • Mahnaz Koupaee, Greg Durrett, Nathanael Chambers, Niranjan Balasubramanian
Event language models represent plausible sequences of events.
no code implementations • Joint Conference on Lexical and Computational Semantics 2021 • Heeyoung Kwon, Nathanael Chambers, Niranjan Balasubramanian
We propose DiP, a Diverse Precondition generation system that can generate unique and diverse preconditions.
1 code implementation • Findings (ACL) 2021 • Yash Kumar Lal, Nathanael Chambers, Raymond Mooney, Niranjan Balasubramanian
They are especially worse on questions whose answers are external to the narrative, thus providing a challenge for future QA and narrative understanding research.
no code implementations • Findings (ACL) 2021 • Tianchu Ji, Shraddhan Jain, Michael Ferdman, Peter Milder, H. Andrew Schwartz, Niranjan Balasubramanian
This informs the design of an inference-time quantization technique using both pruning and log-scaled mapping which produces only a few (e. g. $2^3$) unique values.
1 code implementation • ACL 2021 • Qingqing Cao, Yash Kumar Lal, Harsh Trivedi, Aruna Balasubramanian, Niranjan Balasubramanian
We present IrEne, an interpretable and extensible energy prediction system that accurately predicts the inference energy consumption of a wide range of Transformer-based NLP models.
no code implementations • 10 Dec 2020 • Qingqing Cao, Oriana Riva, Aruna Balasubramanian, Niranjan Balasubramanian
We present a practical approach, called BewQA, that can answer Bew queries by mining a template of the business-related webpages and using the template to guide the search.
1 code implementation • 15 Nov 2020 • Amanpreet Singh, Niranjan Balasubramanian
The dataset introduces a new challenge for summarization in the business domain, requiring highly abstractive and more concise summaries as compared to other existing datasets.
1 code implementation • COLING 2020 • Mohaddeseh Bastan, Mahnaz Koupaee, Youngseo Son, Richard Sicoli, Niranjan Balasubramanian
We introduce PerSenT, a dataset of crowd-sourced annotations of the sentiment expressed by the authors towards the main entities in news articles.
no code implementations • 19 Oct 2020 • Ishita Doshi, Dhritiman Das, Ashish Bhutani, Rajeev Kumar, Rushi Bhatt, Niranjan Balasubramanian
Nearest neighbor search (NNS) has a wide range of applications in information retrieval, computer vision, machine learning, databases, and other areas.
1 code implementation • EMNLP (sustainlp) 2020 • Qingqing Cao, Aruna Balasubramanian, Niranjan Balasubramanian
In this work, we show that existing software-based energy measurements are not accurate because they do not take into account hardware differences and how resource utilization affects energy consumption.
1 code implementation • Findings of the Association for Computational Linguistics 2020 • Heeyoung Kwon, Mahnaz Koupaee, Pratyush Singh, Gargi Sawhney, Anmol Shukla, Keerthi Kumar Kallur, Nathanael Chambers, Niranjan Balasubramanian
This paper introduces PeKo, a crowd-sourced annotation of preconditions between event pairs in newswire, an order of magnitude larger than prior text annotations.
no code implementations • ACL 2020 • Veronica Lynn, Niranjan Balasubramanian, H. Andrew Schwartz
Not all documents are equally important.
1 code implementation • ACL 2020 • Radhika Gaonkar, Heeyoung Kwon, Mohaddeseh Bastan, Niranjan Balasubramanian, Nathanael Chambers
Predicting how events induce emotions in the characters of a story is typically seen as a standard multi-label classification task, which usually treats labels as anonymous classes to predict.
Ranked #1 on
Emotion Classification
on ROCStories
1 code implementation • ACL 2020 • Qingqing Cao, Harsh Trivedi, Aruna Balasubramanian, Niranjan Balasubramanian
It turns out that we can get by without input-wide self-attention at all layers, especially in the lower layers.
1 code implementation • EMNLP 2020 • Harsh Trivedi, Niranjan Balasubramanian, Tushar Khot, Ashish Sabharwal
For a recent large-scale model (XLNet), we show that only 18 points out of its answer F1 score of 72 on HotpotQA are obtained through multifact reasoning, roughly the same as that of a simpler RNN baseline.
1 code implementation • CONLL 2020 • Noah Weber, Leena Shekhar, Heeyoung Kwon, Niranjan Balasubramanian, Nathanael Chambers
A SLDS is a dynamical system in which the latent dynamics of the system (i. e. how the state vector transforms over time) is controlled by top-level discrete switching variables.
no code implementations • 19 Nov 2019 • Yingru Liu, Xuewen Yang, Dongliang Xie, Xin Wang, Li Shen, Hao-Zhi Huang, Niranjan Balasubramanian
In this paper, we propose a novel deep learning model called Task Adaptive Activation Network (TAAN) that can automatically learn the optimal network architecture for MTL.
no code implementations • IJCNLP 2019 • Xuewen Yang, Yingru Liu, Dongliang Xie, Xin Wang, Niranjan Balasubramanian
In this work, we introduce a new latent variable model, LaSyn, that captures the co-dependence between syntax and semantics, while allowing for effective and efficient inference over the latent space.
no code implementations • WS 2019 • Veronica Lynn, Salvatore Giorgi, Niranjan Balasubramanian, H. Andrew Schwartz
NLP naturally puts a primary focus on leveraging document language, occasionally considering user attributes as supplemental.
4 code implementations • NAACL 2019 • Harsh Trivedi, Heeyoung Kwon, Tushar Khot, Ashish Sabharwal, Niranjan Balasubramanian
We introduce Multee, a general architecture that can effectively use entailment models for multi-hop QA tasks.
no code implementations • NAACL 2019 • Jun Seok Kang, Robert L. Logan IV, Zewei Chu, Yang Chen, Dheeru Dua, Kevin Gimpel, Sameer Singh, Niranjan Balasubramanian
Given a sentence about a target entity, the task is to automatically generate a post-modifier phrase that provides contextually relevant information about the entity.
1 code implementation • EMNLP 2018 • Noah Weber, Leena Shekhar, Niranjan Balasubramanian, Nathanael Chambers
This permits the decoder to softly decide what portions of the latent hierarchy to condition on by attending over the value embeddings for a given setting.
no code implementations • EMNLP 2018 • Mohammadzaman Zamani, H. Andrew Schwartz, Veronica E. Lynn, Salvatore Giorgi, Niranjan Balasubramanian
Predictive models over social media language have shown promise in capturing community outcomes, but approaches thus far largely neglect the socio-demographic context (e. g. age, education rates, race) of the community from which the language originates.
no code implementations • ICLR 2019 • Viresh Ranjan, Heeyoung Kwon, Niranjan Balasubramanian, Minh Hoai
We automatically generate fake sentences by corrupting original sentences from a source collection and train the encoders to produce representations that are effective at detecting fake sentences.
2 code implementations • WS 2018 • Noah Weber, Leena Shekhar, Niranjan Balasubramanian
Seq2Seq based neural architectures have become the go-to architecture to apply to sequence to sequence language tasks.
no code implementations • 19 Mar 2018 • Noah Weber, Leena Shekhar, Niranjan Balasubramanian, Kyunghyun Cho
Attention-based neural abstractive summarization systems equipped with copy mechanisms have shown promising results.
1 code implementation • 21 Nov 2017 • Noah Weber, Niranjan Balasubramanian, Nathanael Chambers
Robust and flexible event representations are important to many core areas in language understanding.
no code implementations • EMNLP 2017 • Veronica Lynn, Youngseo Son, Vivek Kulkarni, Niranjan Balasubramanian, H. Andrew Schwartz
We pose the general task of user-factor adaptation {--} adapting supervised learning models to real-valued user factors inferred from a background of their language, reflecting the idea that a piece of text should be understood within the context of the user that wrote it.
1 code implementation • 3 Jun 2017 • Qingqing Cao, Niranjan Balasubramanian, Aruna Balasubramanian
In this paper, we explore optimizations to run Recurrent Neural Network (RNN) models locally on mobile devices.
no code implementations • COLING 2016 • Peter Jansen, Niranjan Balasubramanian, Mihai Surdeanu, Peter Clark
These explanations are used to create a fine-grained categorization of the requirements.
no code implementations • 10 Jul 2015 • Tushar Khot, Niranjan Balasubramanian, Eric Gribkoff, Ashish Sabharwal, Peter Clark, Oren Etzioni
In the first, we simply use the extracted science rules directly as MLN clauses.