no code implementations • 7 Feb 2025 • Lokesh Nagalapatti, Pranava Singhal, Avishek Ghosh, Sunita Sarawagi
Our goal in this paper is to systematically analyze the role of simulators in estimating CATE.
no code implementations • 20 Dec 2024 • Annie D'souza, Swetha M, Sunita Sarawagi
%In this paper, we start with the observation that imbalanced data training of generative models trained imbalanced dataset which under-represent the minority class.
no code implementations • 23 Nov 2024 • Ashwin Ramachandran, Sunita Sarawagi
Calibration is crucial as large language models (LLMs) are increasingly deployed to convert natural language queries into SQL for commercial databases.
1 code implementation • 29 Aug 2024 • Ashish Mittal, Darshan Prabhu, Sunita Sarawagi, Preethi Jyothi
A challenge of our proposed coupling is handling the mismatch between the tokenizers of the LLM and ASR systems.
no code implementations • 3 Jul 2024 • Ashutosh Sathe, Sunita Sarawagi
We introduce an alternative compact representation of the next token distribution that, in expectation, aligns with the complete $n$-gram distribution while markedly reducing variance across mini-batches compared to the standard next-token loss.
1 code implementation • 6 Jun 2024 • Lokesh Nagalapatti, Pranava Singhal, Avishek Ghosh, Sunita Sarawagi
Given a dataset of individuals each described by a covariate vector, a treatment, and an observed outcome on the treatment, the goal of the individual treatment effect (ITE) estimation task is to predict outcome changes resulting from a change in treatment.
1 code implementation • 27 Jan 2024 • Lokesh Nagalapatti, Akshay Iyer, Abir De, Sunita Sarawagi
The main challenge in this estimation task is the potential confounding of treatment assignment with an individual's covariates in the training data, whereas during inference ICTE requires prediction on independently sampled treatments.
1 code implementation • 2 Nov 2023 • Mayank Kothyari, Dhruva Dhingra, Sunita Sarawagi, Soumen Chakrabarti
Standard dense retrieval techniques are inadequate for schema subsetting of a large structured database, where the correct semantics of retrieval demands that we rank sets of schema elements rather than individual elements.
1 code implementation • 20 Oct 2023 • Adithya Bhaskar, Tushar Tomar, Ashutosh Sathe, Sunita Sarawagi
Research in Text-to-SQL conversion has been largely benchmarked against datasets where each text query corresponds to one correct SQL.
no code implementations • 11 Jul 2023 • Vinit S. Unni, Ashish Mittal, Preethi Jyothi, Sunita Sarawagi
RNN-Transducers (RNN-Ts) have gained widespread acceptance as an end-to-end model for speech to text conversion because of their high accuracy and streaming capabilities.
no code implementations • 10 Jan 2023 • Abhijeet Awasthi, Soumen Chakrabarti, Sunita Sarawagi
To the best of our knowledge, we are the first to attempt inference-time adaptation of Text-to-SQL models, and harness trainable structured similarity between subqueries.
no code implementations • 29 Oct 2022 • Abhijeet Awasthi, Ashutosh Sathe, Sunita Sarawagi
Text-to-SQL parsers typically struggle with databases unseen during the train time.
no code implementations • 13 Oct 2022 • Abhijeet Awasthi, Nitish Gupta, Bidisha Samanta, Shachi Dave, Sunita Sarawagi, Partha Talukdar
Despite cross-lingual generalization demonstrated by pre-trained multilingual models, the translate-train paradigm of transferring English datasets across multiple languages remains to be a key mechanism for training task-specific multilingual models.
no code implementations • ACL 2022 • Soumya Chatterjee, Sunita Sarawagi, Preethi Jyothi
Online alignment in machine translation refers to the task of aligning a target word to a source word when the target sequence has only been partially decoded.
1 code implementation • ACL 2022 • Vaidehi Patil, Partha Talukdar, Sunita Sarawagi
This results in improved zero-shot transfer from related HRLs to LRLs without reducing HRL representation and accuracy.
no code implementations • 21 Feb 2022 • Vinit Unni, Shreya Khare, Ashish Mittal, Preethi Jyothi, Sunita Sarawagi, Samarth Bharadwaj
RNN-Transducer (RNN-T) models have become synonymous with streaming end-to-end ASR systems.
1 code implementation • 5 Nov 2021 • Prathamesh Deshpande, Sunita Sarawagi
Long range forecasts are the starting point of many decision support systems that need to draw inference from high-level aggregate patterns on forecasted values.
1 code implementation • ICLR 2022 • Vihari Piratla, Praneeth Netrapalli, Sunita Sarawagi
We consider the problem of training a classification model with group annotated training data.
1 code implementation • NeurIPS 2021 • Anshul Nasery, Soumyadeep Thakur, Vihari Piratla, Abir De, Sunita Sarawagi
In several real world applications, machine learning models are deployed to make predictions on data whose distribution changes gradually along time, leading to a drift between the train and test distributions.
1 code implementation • NeurIPS 2021 • Vihari Piratla, Soumen Chakrabarty, Sunita Sarawagi
Our goal is to evaluate the accuracy of a black-box classification model, not as a single aggregate on a given test data distribution, but as a surface over a large number of combinations of attributes characterizing multiple test data distributions.
1 code implementation • ACL 2021 • Yash Khemchandani, Sarvesh Mehtani, Vaidehi Patil, Abhijeet Awasthi, Partha Talukdar, Sunita Sarawagi
RelateLM uses transliteration to convert the unseen script of limited LRL text into the script of a Related Prominent Language (RPL) (Hindi in our case).
1 code implementation • NAACL 2021 • Abhirut Gupta, Aditya Vavre, Sunita Sarawagi
Machine translation of user-generated code-mixed inputs to English is of crucial importance in applications like web search and targeted advertising.
1 code implementation • 8 Apr 2021 • Arjit Jain, Sunita Sarawagi, Prithviraj Sen
We propose DIAL, a scalable active learning approach that jointly learns embeddings to maximize recall for blocking and accuracy for matching blocked pairs.
1 code implementation • 4 Mar 2021 • Abhijeet Awasthi, Aman Kansal, Sunita Sarawagi, Preethi Jyothi
We consider the task of personalizing ASR models while being constrained by a fixed budget on recording speaker-specific utterances.
1 code implementation • 2 Mar 2021 • Parikshit Bansal, Prathamesh Deshpande, Sunita Sarawagi
Missing values are commonplace in decision support platforms that aggregate data over long time stretches from disparate sources, and reliable data analytics calls for careful handling of missing data.
1 code implementation • 8 Jan 2021 • Prathamesh Deshpande, Kamlesh Marathe, Abir De, Sunita Sarawagi
In recent years, marked temporal point processes (MTPPs) have emerged as a powerful modeling machinery to characterize asynchronous events in a wide variety of applications.
no code implementations • Findings of the Association for Computational Linguistics 2020 • Sahil Shah, Vihari Piratla, Soumen Chakrabarti, Sunita Sarawagi
Each client uses an unsupervised, corpus-based sketch to register to the service.
no code implementations • WS 2020 • Sriram Balasubramanian, Naman jain, Gaurav Jindal, Abhijeet Awasthi, Sunita Sarawagi
We evaluate named entity representations of BERT-based NLP models by investigating their robustness to replacements from the same typed class in the input.
1 code implementation • 24 Jun 2020 • Kartik Khandelwal, Preethi Jyothi, Abhijeet Awasthi, Sunita Sarawagi
Accordingly, we propose a novel coupling of an open-source accent-tuned local model with the black-box service where the output from the service guides frame-level inference in the local model.
2 code implementations • ICLR 2020 • Abhijeet Awasthi, Sabyasachi Ghosh, Rasna Goyal, Sunita Sarawagi
Empirical evaluation on five different tasks shows that (1) our algorithm is more accurate than several existing methods of learning from a mix of clean and noisy supervision, and (2) the coupled rule-exemplar supervision is effective in denoising rules.
2 code implementations • ICML 2020 • Vihari Piratla, Praneeth Netrapalli, Sunita Sarawagi
The domain specific components are discarded after training and only the common component is retained.
Ranked #1 on
Domain Generalization
on LipitK
2 code implementations • 22 Nov 2019 • Oishik Chatterjee, Ganesh Ramakrishnan, Sunita Sarawagi
Scarcity of labeled data is a bottleneck for supervised learning models.
1 code implementation • IJCNLP 2019 • Abhijeet Awasthi, Sunita Sarawagi, Rasna Goyal, Sabyasachi Ghosh, Vihari Piratla
We present a Parallel Iterative Edit (PIE) model for the problem of local sequence transduction arising in tasks like Grammatical error correction (GEC).
Ranked #17 on
Grammatical Error Correction
on CoNLL-2014 Shared Task
1 code implementation • 24 Jun 2019 • Prathamesh Deshpande, Sunita Sarawagi
We present ARU, an Adaptive Recurrent Unit for streaming adaptation of deep globally trained time-series forecasting models.
1 code implementation • ACL 2019 • Vihari Piratla, Sunita Sarawagi, Soumen Chakrabarti
Given a small corpus $\mathcal D_T$ pertaining to a limited set of focused topics, our goal is to train embeddings that accurately capture the sense of words in the topic in spite of the limited size of $\mathcal D_T$.
no code implementations • ICLR 2019 • Shiv Shankar, Sunita Sarawagi
Modern neural architectures critically rely on attention for mapping structured inputs to sequences.
no code implementations • 3 Mar 2019 • Aviral Kumar, Sunita Sarawagi
We study the calibration of several state of the art neural machine translation(NMT) systems built on attention-based encoder-decoder models.
1 code implementation • EMNLP 2018 • Shiv Shankar, Siddhant Garg, Sunita Sarawagi
In this paper we show that a simple beam approximation of the joint distribution between attention and output is an easy, accurate, and efficient attention mechanism for sequence to sequence learning.
1 code implementation • ICML 2018 • Aviral Kumar, Sunita Sarawagi, Ujjwal Jain
Modern neural networks have recently been found to be poorly calibrated, primarily in the direction of over-confidence.
1 code implementation • ICLR 2018 • Shiv Shankar, Vihari Piratla, Soumen Chakrabarti, Siddhartha Chaudhuri, Preethi Jyothi, Sunita Sarawagi
We present CROSSGRAD, a method to use multi-domain training data to learn a classifier that generalizes to new domains.
Ranked #92 on
Domain Generalization
on PACS
no code implementations • 10 Mar 2018 • Srayanta Mukherjee, Devashish Shankar, Atin Ghosh, Nilam Tathawadekar, Pramod Kompalli, Sunita Sarawagi, Krishnendu Chaudhury
Accurate demand forecasts can help on-line retail organizations better plan their supply-chain processes.
no code implementations • 5 Jul 2017 • Shiv Shankar, Sunita Sarawagi
In this paper, we establish their potential in online adapting a batch trained neural network to domain-relevant labeled data at deployment time.
no code implementations • EMNLP 2016 • Pavel Sountsov, Sunita Sarawagi
Encoder-decoder networks are popular for modeling sequences probabilistically in many applications.
no code implementations • 14 May 2016 • Aman Madaan, Sunita Sarawagi
This is owing to the severe mismatch in the distributions of such entities on the web and in the relatively diminutive training data.