Amit Awekar

Rohit Raj Rai, Angana Borah, Amit Awekar

Large Deep Learning models are compressed and deployed for specific applications. However, current Deep Learning model compression methods do not utilize the information about the target application. As a result, the compressed models are application agnostic. Our goal is to customize the model compression process to create a compressed model that will perform better for the target application. Our method, Application Specific Compression (ASC), identifies and prunes components of the large Deep Learning model that are redundant specifically for the given target application. The intuition of our work is to prune the parts of the network that do not contribute significantly to updating the data representation for the given application. We have experimented with the BERT family of models for three applications: Extractive QA, Natural Language Inference, and Paraphrase Identification. We observe that customized compressed models created using ASC method perform better than existing model compression methods and off-the-shelf compressed models.

Akshay Parekh, Ashish Anand, Amit Awekar

The overarching objective of this paper is two-fold. First, to explore model-based approaches to characterize the primary cause of the noise. in the RE dataset TACRED Second, to identify the potentially noisy instances. Towards the first objective, we analyze predictions and performance of state-of-the-art (SOTA) models to identify the root cause of noise in the dataset. Our analysis of TACRED shows that the majority of the noise in the dataset originates from the instances labeled as no-relation which are negative examples. For the second objective, we explore two nearest-neighbor-based strategies to automatically identify potentially noisy examples for elimination and reannotation. Our first strategy, referred to as Intrinsic Strategy (IS), is based on the assumption that positive examples are clean. Thus, we have used false-negative predictions to identify noisy negative examples. Whereas, our second approach, referred to as Extrinsic Strategy, is based on using a clean subset of the dataset to identify potentially noisy negative examples. Finally, we retrained the SOTA models on the eliminated and reannotated dataset. Our empirical results based on two SOTA models trained on TACRED-E following the IS show an average 4% F1-score improvement, whereas reannotation (TACRED-R) does not improve the original results. However, following ES, SOTA models show the average F1-score improvement of 3.8% and 4.4% when trained on respective eliminated (TACRED-EN) and reannotated (TACRED-RN) datasets respectively. We further extended the ES for cleaning positive examples as well, which resulted in an average performance improvement of 5.8% and 5.6% for the eliminated (TACRED-ENP) and reannotated (TACRED-RNP) datasets respectively.

Rohit Raj Rai, Rishant Pal, Amit Awekar

Large neural models are often compressed before deployment. Model compression is necessary for many practical reasons, such as inference latency, memory footprint, and energy consumption. Compressed models are assumed to be miniature versions of corresponding large neural models. However, we question this belief in our work. We compare compressed models with corresponding large neural models using four model characteristics: prediction errors, data representation, data distribution, and vulnerability to adversarial attack. We perform experiments using the BERT-large model and its five compressed versions. For all four model characteristics, compressed models significantly differ from the BERT-large model. Even among compressed models, they differ from each other on all four model characteristics. Apart from the expected loss in model performance, there are major side effects of using compressed models to replace large neural models.

Shashank Aggarwal, Ram Vikas Mishra, Amit Awekar

In multi-agent IR pipelines for tasks such as search and ranking, LLM-based agents exchange intermediate reasoning in terms of Chain-of-Thought (CoT) with each other. Current CoT evaluation narrowly focuses on target task accuracy. However, this metric fails to assess the quality or utility of the reasoning process itself. To address this limitation, we introduce two novel measures: reusability and verifiability. We decouple CoT generation from execution using a Thinker-Executor framework. Reusability measures how easily an Executor can reuse the Thinker's CoT. Verifiability measures how frequently an Executor can match the Thinker's answer using the CoT. We evaluated four Thinker models against a committee of ten Executor models across five benchmarks. Our results reveal that reusability and verifiability do not correlate with standard accuracy, exposing a blind spot in current accuracy-based leaderboards for reasoning capability. Surprisingly, we find that CoTs from specialized reasoning models are not consistently more reusable or verifiable than those from general-purpose LLMs like Llama and Gemma.

Rohit Raj Rai, Abhishek Dhaka, Amit Awekar

Compressed Deep Learning (DL) models are essential for deployment in resource-constrained environments. But their performance often lags behind their large-scale counterparts. To bridge this gap, we propose Alignment Adapter (AlAd): a lightweight, sliding-window-based adapter. It aligns the token-level embeddings of a compressed model with those of the original large model. AlAd preserves local contextual semantics, enables flexible alignment across differing dimensionalities or architectures, and is entirely agnostic to the underlying compression method. AlAd can be deployed in two ways: as a plug-and-play module over a frozen compressed model, or by jointly fine-tuning AlAd with the compressed model for further performance gains. Through experiments on BERT-family models across three token-level NLP tasks, we demonstrate that AlAd significantly boosts the performance of compressed models with only marginal overhead in size and latency.

Rohit Raj Rai, Chirag Kothari, Siddhesh Shelke et al.

Large Deep Learning models are often compressed before being deployed in a resource-constrained environment. Can we trust the prediction of compressed models just as we trust the prediction of the original large model? Existing work has keenly studied the effect of compression on accuracy and related performance measures. However, performance parity does not guarantee trust-equivalence. We propose a two-dimensional framework for trust-equivalence evaluation. First, interpretability alignment measures whether the models base their predictions on the same input features. We use LIME and SHAP tests to measure the interpretability alignment. Second, calibration similarity measures whether the models exhibit comparable reliability in their predicted probabilities. It is assessed via ECE, MCE, Brier Score, and reliability diagrams. We conducted experiments using BERT-base as the large model and its multiple compressed variants. We focused on two text classification tasks: natural language inference and paraphrase identification. Our results reveal low interpretability alignment and significant mismatch in calibration similarity. It happens even when the accuracies are nearly identical between models. These findings show that compressed models are not trust-equivalent to their large counterparts. Deploying compressed models as a drop-in replacement for large models requires careful assessment, going beyond performance parity.

Rohit Raj Rai, Amit Awekar

Word embedding methods (WEMs) are extensively used for representing text data. The dimensionality of these embeddings varies across various tasks and implementations. The effect of dimensionality change on the accuracy of the downstream task is a well-explored question. However, how the dimensionality change affects the bias of word embeddings needs to be investigated. Using the English Wikipedia corpus, we study this effect for two static (Word2Vec and fastText) and two context-sensitive (ElMo and BERT) WEMs. We have two observations. First, there is a significant variation in the bias of word embeddings with the dimensionality change. Second, there is no uniformity in how the dimensionality change affects the bias of word embeddings. These factors should be considered while selecting the dimensionality of word embeddings.

Akshay Parekh, Ashish Anand, Amit Awekar

Large crowd-sourced datasets are often noisy and relation classification (RC) datasets are no exception. Reannotating the entire dataset is one probable solution however it is not always viable due to time and budget constraints. This paper addresses the problem of efficient reannotation of a large noisy dataset for the RC. Our goal is to catch more annotation errors in the dataset while reannotating fewer instances. Existing work on RC dataset reannotation lacks the flexibility about how much data to reannotate. We introduce the concept of a reannotation budget to overcome this limitation. The immediate follow-up problem is: Given a specific reannotation budget, which subset of the data should we reannotate? To address this problem, we present two strategies to selectively reannotate RC datasets. Our strategies utilize the taxonomic hierarchy of relation labels. The intuition of our work is to rely on the graph distance between actual and predicted relation labels in the label hierarchy graph. We evaluate our reannotation strategies on the well-known TACRED dataset. We design our experiments to answer three specific research questions. First, does our strategy select novel candidates for reannotation? Second, for a given reannotation budget is our reannotation strategy more efficient at catching annotation errors? Third, what is the impact of data reannotation on RC model performance measurement? Experimental results show that our both reannotation strategies are novel and efficient. Our analysis indicates that the current reported performance of RC models on noisy TACRED data is inflated.

Angana Borah, Manash Pratim Barman, Amit Awekar

A representation learning method is considered stable if it consistently generates similar representation of the given data across multiple runs. Word Embedding Methods (WEMs) are a class of representation learning methods that generate dense vector representation for each word in the given text data. The central idea of this paper is to explore the stability measurement of WEMs using intrinsic evaluation based on word similarity. We experiment with three popular WEMs: Word2Vec, GloVe, and fastText. For stability measurement, we investigate the effect of five parameters involved in training these models. We perform experiments using four real-world datasets from different domains: Wikipedia, News, Song lyrics, and European parliament proceedings. We also observe the effect of WEM stability on three downstream tasks: Clustering, POS tagging, and Fairness evaluation. Our experiments indicate that amongst the three WEMs, fastText is the most stable, followed by GloVe and Word2Vec.

Akshay Parekh, Ashish Anand, Amit Awekar

This work addresses two important questions pertinent to Relation Extraction (RE). First, what are all possible relations that could exist between any two given entity types? Second, how do we define an unambiguous taxonomical (is-a) hierarchy among the identified relations? To address the first question, we use three resources Wikipedia Infobox, Wikidata, and DBpedia. This study focuses on relations between person, organization and location entity types. We exploit Wikidata and DBpedia in a data-driven manner, and Wikipedia Infobox templates manually to generate lists of relations. Further, to address the second question, we canonicalize, filter, and combine the identified relations from the three resources to construct a taxonomical hierarchy. This hierarchy contains 623 canonical relations with highest contribution from Wikipedia Infobox followed by DBpedia and Wikidata. The generated relation list subsumes an average of 85% of relations from RE datasets when entity types are restricted.

Manash Pratim Barman, Amit Awekar, Sambhav Kothari

The central idea of this paper is to gain a deeper understanding of song lyrics computationally. We focus on two aspects: style and biases of song lyrics. All prior works to understand these two aspects are limited to manual analysis of a small corpus of song lyrics. In contrast, we analyzed more than half a million songs spread over five decades. We characterize the lyrics style in terms of vocabulary, length, repetitiveness, speed, and readability. We have observed that the style of popular songs significantly differs from other songs. We have used distributed representation methods and WEAT test to measure various gender and racial biases in the song lyrics. We have observed that biases in song lyrics correlate with prior results on human subjects. This correlation indicates that song lyrics reflect the biases that exist in society. Increasing consumption of music and the effect of lyrics on human emotions makes this analysis important.

Abhishek Abhishek, Sanya Bathla Taneja, Garima Malik et al.

Fine-grained Entity Recognition (FgER) is the task of detecting and classifying entity mentions to a large set of types spanning diverse domains such as biomedical, finance and sports. We observe that when the type set spans several domains, detection of entity mention becomes a limitation for supervised learning models. The primary reason being lack of dataset where entity boundaries are properly annotated while covering a large spectrum of entity types. Our work directly addresses this issue. We propose Heuristics Allied with Distant Supervision (HAnDS) framework to automatically construct a quality dataset suitable for the FgER task. HAnDS framework exploits the high interlink among Wikipedia and Freebase in a pipelined manner, reducing annotation errors introduced by naively using distant supervision approach. Using HAnDS framework, we create two datasets, one suitable for building FgER systems recognizing up to 118 entity types based on the FIGER type hierarchy and another for up to 1115 entity types based on the TypeNet hierarchy. Our extensive empirical experimentation warrants the quality of the generated datasets. Along with this, we also provide a manually annotated dataset for benchmarking FgER systems.

Manash Pratim Barman, Kavish Dahekar, Abhinav Anshuman et al.

The central idea of this paper is to demonstrate the strength of lyrics for music mining and natural language processing (NLP) tasks using the distributed representation paradigm. For music mining, we address two prediction tasks for songs: genre and popularity. Existing works for both these problems have two major bottlenecks. First, they represent lyrics using handcrafted features that require intricate knowledge of language and music. Second, they consider lyrics as a weak indicator of genre and popularity. We overcome both the bottlenecks by representing lyrics using distributed representation. In our work, genre identification is a multi-class classification task whereas popularity prediction is a binary classification task. We achieve an F1 score of around 0.6 for both the tasks using only lyrics. Distributed representation of words is now heavily used for various NLP algorithms. We show that lyrics can be used to improve the quality of this representation.

Abhishek Abhishek, Amar Prakash Azad, Balaji Ganesan et al.

Entity typing (ET) is the problem of assigning labels to given entity mentions in a sentence. Existing works for ET require knowledge about the domain and target label set for a given test instance. ET in the absence of such knowledge is a novel problem that we address as ET in the wild. We hypothesize that the solution to this problem is to build supervised models that generalize better on the ET task as a whole, rather than a specific dataset. In this direction, we propose a Collective Learning Framework (CLF), which enables learning from diverse datasets in a unified way. The CLF first creates a unified hierarchical label set (UHLS) and a label mapping by aggregating label information from all available datasets. Then it builds a single neural network classifier using UHLS, label mapping, and a partial loss function. The single classifier predicts the finest possible label across all available domains even though these labels may not be present in any domain-specific dataset. We also propose a set of evaluation schemes and metrics to evaluate the performance of models in this novel problem. Extensive experimentation on seven diverse real-world datasets demonstrates the efficacy of our CLF.

Sweta Agrawal, Amit Awekar

Harassment by cyberbullies is a significant phenomenon on the social media. Existing works for cyberbullying detection have at least one of the following three bottlenecks. First, they target only one particular social media platform (SMP). Second, they address just one topic of cyberbullying. Third, they rely on carefully handcrafted features of the data. We show that deep learning based models can overcome all three bottlenecks. Knowledge learned by these models on one dataset can be transferred to other datasets. We performed extensive experiments using three real-world datasets: Formspring (12k posts), Twitter (16k posts), and Wikipedia(100k posts). Our experiments provide several useful insights about cyberbullying detection. To the best of our knowledge, this is the first work that systematically analyzes cyberbullying detection on various topics across multiple SMPs using deep learning based models and transfer learning.

Abhishek, Ashish Anand, Amit Awekar

Fine-grained entity type classification (FETC) is the task of classifying an entity mention to a broad set of types. Distant supervision paradigm is extensively used to generate training data for this task. However, generated training data assigns same set of labels to every mention of an entity without considering its local context. Existing FETC systems have two major drawbacks: assuming training data to be noise free and use of hand crafted features. Our work overcomes both drawbacks. We propose a neural network model that jointly learns entity mentions and their context representation to eliminate use of hand crafted features. Our model treats training data as noisy and uses non-parametric variant of hinge loss function. Experiments show that the proposed model outperforms previous state-of-the-art methods on two publicly available datasets, namely FIGER (GOLD) and BBN with an average relative improvement of 2.69% in micro-F1 score. Knowledge learnt by our model on one dataset can be transferred to other datasets while using same model or other FETC systems. These approaches of transferring knowledge further improve the performance of respective models.

Panthadeep Bhattacharjee, Amit Awekar

Incremental data mining algorithms process frequent updates to dynamic datasets efficiently by avoiding redundant computation. Existing incremental extension to shared nearest neighbor density based clustering (SNND) algorithm cannot handle deletions to dataset and handles insertions only one point at a time. We present an incremental algorithm to overcome both these bottlenecks by efficiently identifying affected parts of clusters while processing updates to dataset in batch mode. We show effectiveness of our algorithm by performing experiments on large synthetic as well as real world datasets. Our algorithm is up to four orders of magnitude faster than SNND and requires up to 60% extra memory than SNND while providing output identical to SNND.

Siddhesh Khandelwal, Amit Awekar

There has been considerable work on improving popular clustering algorithm `K-means' in terms of mean squared error (MSE) and speed, both. However, most of the k-means variants tend to compute distance of each data point to each cluster centroid for every iteration. We propose a fast heuristic to overcome this bottleneck with only marginal increase in MSE. We observe that across all iterations of K-means, a data point changes its membership only among a small subset of clusters. Our heuristic predicts such clusters for each data point by looking at nearby clusters after the first iteration of k-means. We augment well known variants of k-means with our heuristic to demonstrate effectiveness of our heuristic. For various synthetic and real-world datasets, our heuristic achieves speed-up of up-to 3 times when compared to efficient variants of k-means.

Anasua Mitra, Amit Awekar

Number of published scholarly articles is growing exponentially. To tackle this information overload, researchers are increasingly depending on niche academic search engines. Recent works have shown that two major general web search engines: Google and Bing, have high level of agreement in their top search results. In contrast, we show that various academic search engines have low degree of agreement among themselves. We performed experiments using 2500 queries over four academic search engines. We observe that overlap in search result sets of any pair of academic search engines is significantly low and in most of the cases the search result sets are mutually exclusive. We also discuss implications of this low overlap.