Ishan Jindal
YOU?
Author Swipe
View article: Real-Time Cooked Food Image Synthesis and Visual Cooking Progress Monitoring on Edge Devices
Real-Time Cooked Food Image Synthesis and Visual Cooking Progress Monitoring on Edge Devices Open
Synthesizing realistic cooked food images from raw inputs on edge devices is a challenging generative task, requiring models to capture complex changes in texture, color and structure during cooking. Existing image-to-image generation meth…
View article: Identifying Noise in Human-Created Datasets using Training Dynamics from Generative Models
Identifying Noise in Human-Created Datasets using Training Dynamics from Generative Models Open
View article: Beyond Labels: Empowering Human Annotators with Natural Language Explanations through a Novel Active-Learning Architecture
Beyond Labels: Empowering Human Annotators with Natural Language Explanations through a Novel Active-Learning Architecture Open
Real-world domain experts (e.g., doctors) rarely annotate only a decision label in their day-to-day workflow without providing explanations. Yet, existing low-resource learning techniques, such as Active Learning (AL), that aim to support …
View article: NL-Augmenter 🦎 → 🐍 A Framework for Task-Sensitive Natural Language Augmentation
NL-Augmenter 🦎 → 🐍 A Framework for Task-Sensitive Natural Language Augmentation Open
Data augmentation is an important method for evaluating the robustness of and enhancing the diversity of training data for natural language processing (NLP) models. In this paper, we present NL-Augmenter, a new participatory Python-based n…
View article: When to Use What: An In-Depth Comparative Empirical Analysis of OpenIE Systems for Downstream Applications
When to Use What: An In-Depth Comparative Empirical Analysis of OpenIE Systems for Downstream Applications Open
Open Information Extraction (OpenIE) has been used in the pipelines of various NLP tasks. Unfortunately, there is no clear consensus on which models to use in which tasks. Muddying things further is the lack of comparisons that take differ…
View article: Abstractive Open Information Extraction
Abstractive Open Information Extraction Open
Open Information Extraction (OpenIE) is a traditional NLP task that extracts structured information from unstructured text to be used for other downstream applications. Traditionally, OpenIE focuses on extracting the surface forms of relat…
View article: Beyond Labels: Empowering Human Annotators with Natural Language Explanations through a Novel Active-Learning Architecture
Beyond Labels: Empowering Human Annotators with Natural Language Explanations through a Novel Active-Learning Architecture Open
Bingsheng Yao, Ishan Jindal, Lucian Popa, Yannis Katsis, Sayan Ghosh, Lihong He, Yuxuan Lu, Shashank Srivastava, Yunyao Li, James Hendler, Dakuo Wang. Findings of the Association for Computational Linguistics: EMNLP 2023. 2023.
View article: PriMeSRL-Eval: A Practical Quality Metric for Semantic Role Labeling Systems Evaluation
PriMeSRL-Eval: A Practical Quality Metric for Semantic Role Labeling Systems Evaluation Open
Semantic role labeling (SRL) identifies the predicate-argument structure in a sentence. This task is usually accomplished in four steps: predicate identification, predicate sense disambiguation, argument identification, and argument classi…
View article: When to Use What: An In-Depth Comparative Empirical Analysis of OpenIE Systems for Downstream Applications
When to Use What: An In-Depth Comparative Empirical Analysis of OpenIE Systems for Downstream Applications Open
Open Information Extraction (OpenIE) has been used in the pipelines of various NLP tasks. Unfortunately, there is no clear consensus on which models to use in which tasks. Muddying things further is the lack of comparisons that take differ…
View article: PriMeSRL-Eval: A Practical Quality Metric for Semantic Role Labeling Systems Evaluation
PriMeSRL-Eval: A Practical Quality Metric for Semantic Role Labeling Systems Evaluation Open
Semantic role labeling (SRL) identifies the predicate-argument structure in a sentence. This task is usually accomplished in four steps: predicate identification, predicate sense disambiguation, argument identification, and argument classi…
View article: Meaning Representations for Natural Languages: Design, Models and Applications
Meaning Representations for Natural Languages: Design, Models and Applications Open
This tutorial reviews the design of common meaning representations, SoTA models for predicting meaning representations, and the applications of meaning representations in a wide range of downstream NLP tasks and real-world applications. Re…
View article: Label Definitions Improve Semantic Role Labeling
Label Definitions Improve Semantic Role Labeling Open
Argument classification is at the core of Semantic Role Labeling. Given a sentence and the predicate, a semantic role label is assigned to each argument of the predicate. While semantic roles come with meaningful definitions, existing work…
View article: NL-Augmenter: A Framework for Task-Sensitive Natural Language\n Augmentation
NL-Augmenter: A Framework for Task-Sensitive Natural Language\n Augmentation Open
Data augmentation is an important component in the robustness evaluation of\nmodels in natural language processing (NLP) and in enhancing the diversity of\nthe data they are trained on. In this paper, we present NL-Augmenter, a new\npartic…
View article: OPAD: An Optimized Policy-based Active Learning Framework for Document Content Analysis
OPAD: An Optimized Policy-based Active Learning Framework for Document Content Analysis Open
Documents are central to many business systems, and include forms, reports, contracts, invoices or purchase orders. The information in documents is typically in natural language, but can be organized in various layouts and formats. There h…
View article: Improved Semantic Role Labeling using Parameterized Neighborhood Memory Adaptation
Improved Semantic Role Labeling using Parameterized Neighborhood Memory Adaptation Open
Deep neural models achieve some of the best results for semantic role labeling. Inspired by instance-based learning that utilizes nearest neighbors to handle low-frequency context-specific training samples, we investigate the use of memory…
View article: CLAR: A Cross-Lingual Argument Regularizer for Semantic Role Labeling
CLAR: A Cross-Lingual Argument Regularizer for Semantic Role Labeling Open
Semantic role labeling (SRL) identifies predicate-argument structure(s) in a given sentence. Although different languages have different argument annotations, polyglot training, the idea of training one model on multiple languages, has pre…
View article: CLAR: A Cross-Lingual Argument Regularizer for Semantic Role Labeling
CLAR: A Cross-Lingual Argument Regularizer for Semantic Role Labeling Open
Semantic role labeling (SRL) identifies predicate-argument structure(s) in a given sentence. Although different languages have different argument annotations, polyglot training, the idea of training one model on multiple languages, has pre…
View article: An Effective Label Noise Model for DNN Text Classification
An Effective Label Noise Model for DNN Text Classification Open
Because large, human-annotated datasets suffer from labeling errors, it is crucial to be able to train deep neural networks in the presence of label noise. While training image classification models with label noise have received much atte…
View article: An Effective Label Noise Model for
An Effective Label Noise Model for Open
Ishan Jindal, Daniel Pressel, Brian Lester, Matthew Nokleby. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers). …
View article: Learning Models For Corrupted Multi-Dimensional Data: Fundamental Limits And Algorithms
Learning Models For Corrupted Multi-Dimensional Data: Fundamental Limits And Algorithms Open
Developing machine learning models for unstructured multi-dimensional datasets such as datasets with unreliable labels and noisy multi-dimensional signals with or without missing information have becoming a central necessity. We are not al…
View article: Optimizing Taxi Carpool Policies via Reinforcement Learning and Spatio-Temporal Mining
Optimizing Taxi Carpool Policies via Reinforcement Learning and Spatio-Temporal Mining Open
In this paper, we develop a reinforcement learning (RL) based system to learn an effective policy for carpooling that maximizes transportation efficiency so that fewer cars are required to fulfill the given amount of trip demand. For this …
View article: Tensor Matched Kronecker-Structured Subspace Detection for Missing\n Information
Tensor Matched Kronecker-Structured Subspace Detection for Missing\n Information Open
We consider the problem of detecting whether a tensor signal having many\nmissing entities lies within a given low dimensional Kronecker-Structured (KS)\nsubspace. This is a matched subspace detection problem. Tensor matched subspace\ndete…
View article: Tensor Matched Kronecker-Structured Subspace Detection for Missing Information
Tensor Matched Kronecker-Structured Subspace Detection for Missing Information Open
We consider the problem of detecting whether a tensor signal having many missing entities lies within a given low dimensional Kronecker-Structured (KS) subspace. This is a matched subspace detection problem. Tensor matched subspace detecti…
View article: Classification and Representation via Separable Subspaces: Performance Limits and Algorithms
Classification and Representation via Separable Subspaces: Performance Limits and Algorithms Open
We study the classification performance of Kronecker-structured models in two\nasymptotic regimes and developed an algorithm for separable, fast and compact\nK-S dictionary learning for better classification and representation of\nmultidim…
View article: A Unified Neural Network Approach for Estimating Travel Time and Distance for a Taxi Trip
A Unified Neural Network Approach for Estimating Travel Time and Distance for a Taxi Trip Open
In building intelligent transportation systems such as taxi or rideshare services, accurate prediction of travel time and distance is crucial for customer experience and resource management. Using the NYC taxi dataset, which contains taxi …
View article: Learning Deep Networks from Noisy Labels with Dropout Regularization
Learning Deep Networks from Noisy Labels with Dropout Regularization Open
Large datasets often have unreliable labels-such as those obtained from Amazon's Mechanical Turk or social media platforms-and classifiers trained on mislabeled datasets often exhibit poor performance. We present a simple, effective techni…
View article: Effective object tracking in unstructured crowd scenes
Effective object tracking in unstructured crowd scenes Open
In this paper, we are presenting a rotation variant Oriented Texture Curve (OTC) descriptor based mean shift algorithm for tracking an object in an unstructured crowd scene. The proposed algorithm works by first obtaining the OTC features …
View article: SA-CNN: Dynamic Scene Classification using Convolutional Neural Networks
SA-CNN: Dynamic Scene Classification using Convolutional Neural Networks Open
The task of classifying videos of natural dynamic scenes into appropriate classes has gained lot of attention in recent years. The problem especially becomes challenging when the camera used to capture the video is dynamic. In this paper, …