Avinash Madasu

Avinash Madasu, Vasudev Lal

Video retrieval (VR) involves retrieving the ground truth video from the video database given a text caption or vice-versa. The two important components of compositionality: objects & attributes and actions are joined using correct syntax to form a proper text query. These components (objects & attributes, actions and syntax) each play an important role to help distinguish among videos and retrieve the correct ground truth video. However, it is unclear what is the effect of these components on the video retrieval performance. We therefore, conduct a systematic study to evaluate the compositional and syntactic understanding of video retrieval models on standard benchmarks such as MSRVTT, MSVD and DIDEMO. The study is performed on two categories of video retrieval models: (i) which are pre-trained on video-text pairs and fine-tuned on downstream video retrieval datasets (Eg. Frozen-in-Time, Violet, MCQ etc.) (ii) which adapt pre-trained image-text representations like CLIP for video retrieval (Eg. CLIP4Clip, XCLIP, CLIP2Video etc.). Our experiments reveal that actions and syntax play a minor role compared to objects & attributes in video understanding. Moreover, video retrieval models that use pre-trained image-text representations (CLIP) have better syntactic and compositional understanding as compared to models pre-trained on video-text data. The code is available at https://github.com/IntelLabs/multimodal_cognitive_ai/tree/main/ICSVR

Avinash Madasu, Junier Oliva, Gedas Bertasius

The majority of traditional text-to-video retrieval systems operate in static environments, i.e., there is no interaction between the user and the agent beyond the initial textual query provided by the user. This can be sub-optimal if the initial query has ambiguities, which would lead to many falsely retrieved videos. To overcome this limitation, we propose a novel framework for Video Retrieval using Dialog (ViReD), which enables the user to interact with an AI agent via multiple rounds of dialog, where the user refines retrieved results by answering questions generated by an AI agent. Our novel multimodal question generator learns to ask questions that maximize the subsequent video retrieval performance using (i) the video candidates retrieved during the last round of interaction with the user and (ii) the text-based dialog history documenting all previous interactions, to generate questions that incorporate both visual and linguistic cues relevant to video retrieval. Furthermore, to generate maximally informative questions, we propose an Information-Guided Supervision (IGS), which guides the question generator to ask questions that would boost subsequent video retrieval accuracy. We validate the effectiveness of our interactive ViReD framework on the AVSD dataset, showing that our interactive method performs significantly better than traditional non-interactive video retrieval systems. We also demonstrate that our proposed approach generalizes to the real-world settings that involve interactions with real humans, thus, demonstrating the robustness and generality of our framework

Phillip Howard, Avinash Madasu, Tiep Le et al.

While vision-language models (VLMs) have achieved remarkable performance improvements recently, there is growing evidence that these models also posses harmful biases with respect to social attributes such as gender and race. Prior studies have primarily focused on probing such bias attributes individually while ignoring biases associated with intersections between social attributes. This could be due to the difficulty of collecting an exhaustive set of image-text pairs for various combinations of social attributes. To address this challenge, we employ text-to-image diffusion models to produce counterfactual examples for probing intersectional social biases at scale. Our approach utilizes Stable Diffusion with cross attention control to produce sets of counterfactual image-text pairs that are highly similar in their depiction of a subject (e.g., a given occupation) while differing only in their depiction of intersectional social attributes (e.g., race & gender). Through our over-generate-then-filter methodology, we produce SocialCounterfactuals, a high-quality dataset containing 171k image-text pairs for probing intersectional biases related to gender, race, and physical characteristics. We conduct extensive experiments to demonstrate the usefulness of our generated dataset for probing and mitigating intersectional social biases in state-of-the-art VLMs.

Avinash Madasu, Estelle Aflalo, Gabriela Ben Melech Stan et al.

Multi-modal retrieval has seen tremendous progress with the development of vision-language models. However, further improving these models require additional labelled data which is a huge manual effort. In this paper, we propose a framework MuMUR, that utilizes knowledge transfer from a multilingual model to boost the performance of multi-modal (image and video) retrieval. We first use state-of-the-art machine translation models to construct pseudo ground-truth multilingual visual-text pairs. We then use this data to learn a joint vision-text representation where English and non-English text queries are represented in a common embedding space based on pretrained multilingual models. We evaluate our proposed approach on a diverse set of retrieval datasets: five video retrieval datasets such as MSRVTT, MSVD, DiDeMo, Charades and MSRVTT multilingual, two image retrieval datasets such as Flickr30k and Multi30k . Experimental results demonstrate that our approach achieves state-of-the-art results on all video retrieval datasets outperforming previous models. Additionally, our framework MuMUR significantly beats other multilingual video retrieval dataset. We also observe that MuMUR exhibits strong performance on image retrieval. This demonstrates the universal ability of MuMUR to perform retrieval across all visual inputs (image and video) and text inputs (monolingual and multilingual).

Avinash Madasu, Shashank Srivastava

Large language models (LMs) have rapidly become a mainstay in Natural Language Processing. These models are known to acquire rich linguistic knowledge from training on large amounts of text. In this paper, we investigate if pre-training on text also confers these models with helpful `inductive biases' for non-linguistic reasoning. On a set of 19 diverse non-linguistic tasks involving quantitative computations, recognizing regular expressions and reasoning over strings. We find that pretrained models significantly outperform comparable non-pretrained neural models. This remains true also in experiments with training non-pretrained models with fewer parameters to account for model regularization effects. We further explore the effect of text domain on LMs by pretraining models from text from different domains and provenances. Our experiments surprisingly reveal that the positive effects of pre-training persist even when pretraining on multi-lingual text or computer code, and even for text generated from synthetic languages. Our findings suggest a hitherto unexplored deep connection between pre-training and inductive learning abilities of language models.

Kilichbek Haydarov, Xiaoqian Shen, Avinash Madasu et al.

We introduce Affective Visual Dialog, an emotion explanation and reasoning task as a testbed for research on understanding the formation of emotions in visually grounded conversations. The task involves three skills: (1) Dialog-based Question Answering (2) Dialog-based Emotion Prediction and (3) Affective emotion explanation generation based on the dialog. Our key contribution is the collection of a large-scale dataset, dubbed AffectVisDial, consisting of 50K 10-turn visually grounded dialogs as well as concluding emotion attributions and dialog-informed textual emotion explanations, resulting in a total of 27,180 working hours. We explain our design decisions in collecting the dataset and introduce the questioner and answerer tasks that are associated with the participants in the conversation. We train and demonstrate solid Affective Visual Dialog baselines adapted from state-of-the-art models. Remarkably, the responses generated by our models show promising emotional reasoning abilities in response to visually grounded conversations. Our project page is available at https://affective-visual-dialog.github.io.

Avinash Madasu, Mauajama Firdaus, Asif Eqbal

Social chatbots have gained immense popularity, and their appeal lies not just in their capacity to respond to the diverse requests from users, but also in the ability to develop an emotional connection with users. To further develop and promote social chatbots, we need to concentrate on increasing user interaction and take into account both the intellectual and emotional quotient in the conversational agents. In this paper, we propose a multi-task framework that jointly identifies the emotion of a given dialogue and generates response in accordance to the identified emotion. We employ a BERT based network for creating an empathetic system and use a mixed objective function that trains the end-to-end network with both the classification and generation loss. Experimental results show that our proposed framework outperforms current state-of-the-art models

Phillip Howard, Avinash Madasu, Tiep Le et al.

While vision-language models (VLMs) have achieved remarkable performance improvements recently, there is growing evidence that these models also posses harmful biases with respect to social attributes such as gender and race. Prior studies have primarily focused on probing such bias attributes individually while ignoring biases associated with intersections between social attributes. This could be due to the difficulty of collecting an exhaustive set of image-text pairs for various combinations of social attributes from existing datasets. To address this challenge, we employ text-to-image diffusion models to produce counterfactual examples for probing intserctional social biases at scale. Our approach utilizes Stable Diffusion with cross attention control to produce sets of counterfactual image-text pairs that are highly similar in their depiction of a subject (e.g., a given occupation) while differing only in their depiction of intersectional social attributes (e.g., race & gender). We conduct extensive experiments using our generated dataset which reveal the intersectional social biases present in state-of-the-art VLMs.

Avinash Madasu, Vasudev Lal

Vision (image and video) - Language (VL) pre-training is the recent popular paradigm that achieved state-of-the-art results on multi-modal tasks like image-retrieval, video-retrieval, visual question answering etc. These models are trained in an unsupervised way and greatly benefit from the complementary modality supervision. In this paper, we explore if the language representations trained using vision supervision perform better than vanilla language representations on Natural Language Understanding and commonsense reasoning benchmarks. We experiment with a diverse set of image-text models such as ALBEF, BLIP, METER and video-text models like ALPRO, Frozen-in-Time (FiT), VIOLET. We compare the performance of language representations of stand-alone text encoders of these models to the language representations of text encoders learnt through vision supervision. Our experiments suggest that vanilla language representations show superior performance on most of the tasks. These results shed light on the current drawbacks of the vision-language models.

Avinash Madasu, Yossi Gandelsman, Vasudev Lal et al.

CLIP is one of the most popular foundational models and is heavily used for many vision-language tasks. However, little is known about the inner workings of CLIP. To bridge this gap we propose a study to quantify the interpretability in CLIP like models. We conduct this study on six different CLIP models from OpenAI and OpenCLIP which vary by size, type of pre-training data and patch size. Our approach begins with using the TEXTSPAN algorithm and in-context learning to break down individual attention heads into specific properties. We then evaluate how easily these heads can be interpreted using new metrics which measure property consistency within heads and property disentanglement across heads. Our findings reveal that larger CLIP models are generally more interpretable than their smaller counterparts. To further assist users in understanding the inner workings of CLIP models, we introduce CLIP-InterpreT, a tool designed for interpretability analysis. CLIP-InterpreT offers five types of analyses: property-based nearest neighbor search, per-head topic segmentation, contrastive segmentation, per-head nearest neighbors of an image, and per-head nearest neighbors of text.

Avinash Madasu, Anahita Bhiwandiwalla, Vasudev Lal

Foundational multimodal models pre-trained on large scale image-text pairs or video-text pairs or both have shown strong generalization abilities on downstream tasks. However unlike image-text models, pretraining video-text models is always not feasible due to the difficulty in collecting large-scale clean and aligned data, and exponential computational costs involved in the pretraining phase. Therefore, the pertinent question to ask is: Can image-text models be adapted to video tasks and is there any benefit to using these models over pretraining directly on videos? In this work, we focus on this question by proposing a detailed study on the generalization abilities of image-text models when evaluated on video understanding tasks in a zero-shot setting. We investigate 9 foundational image-text models on a diverse set of video tasks that include video action recognition (video AR), video retrieval (video RT), video question answering (video QA), video multiple choice (video MC) and video captioning (video CP). Our experiments show that image-text models exhibit impressive performance on video AR, video RT and video MC. Furthermore, they perform moderately on video captioning and poorly on video QA. These findings shed a light on the benefits of adapting foundational image-text models to an array of video tasks while avoiding the costly pretraining step.

Avinash Madasu, Anvesh Rao Vijjini

A well formed query is defined as a query which is formulated in the manner of an inquiry, and with correct interrogatives, spelling and grammar. While identifying well formed queries is an important task, few works have attempted to address it. In this paper we propose transformer based language model - Bidirectional Encoder Representations from Transformers (BERT) to this task. We further imbibe BERT with parts-of-speech information inspired from earlier works. Furthermore, we also train the model in multiple curriculum settings for improvement in performance. Curriculum Learning over the task is experimented with Baby Steps and One Pass techniques. Proposed architecture performs exceedingly well on the task. The best approach achieves accuracy of 83.93%, outperforming previous state-of-the-art at 75.0% and reaching close to the approximate human upper bound of 88.4%.

Mauajama Firdaus, Avinash Madasu, Asif Ekbal

Natural Language Understanding (NLU) and Natural Language Generation (NLG) are the two critical components of every conversational system that handles the task of understanding the user by capturing the necessary information in the form of slots and generating an appropriate response in accordance with the extracted information. Recently, dialogue systems integrated with complementary information such as images, audio, or video have gained immense popularity. In this work, we propose an end-to-end framework with the capability to extract necessary slot values from the utterance and generate a coherent response, thereby assisting the user to achieve their desired goals in a multimodal dialogue system having both textual and visual information. The task of extracting the necessary information is dependent not only on the text but also on the visual cues present in the dialogue. Similarly, for the generation, the previous dialog context comprising multimodal information is significant for providing coherent and informative responses. We employ a multimodal hierarchical encoder using pre-trained DialoGPT and also exploit the knowledge base (Kb) to provide a stronger context for both the tasks. Finally, we design a slot attention mechanism to focus on the necessary information in a given utterance. Lastly, a decoder generates the corresponding response for the given dialogue context and the extracted slot values. Experimental results on the Multimodal Dialogue Dataset (MMD) show that the proposed framework outperforms the baselines approaches in both the tasks. The code is available at https://github.com/avinashsai/slot-gpt.

Avinash Madasu, Vasudev Lal, Phillip Howard

Vision-Language Models (VLMs) are increasingly deployed in diverse cultural contexts, yet their internal biases remain poorly understood. In this work, we propose a novel framework to systematically evaluate how VLMs encode cultural differences and biases related to race, gender, and physical traits across countries. We introduce three retrieval-based tasks: (1) Race to Country retrieval, which examines the association between individuals from specific racial groups (East Asian, White, Middle Eastern, Latino, South Asian, and Black) and different countries; (2) Personal Traits to Country retrieval, where images are paired with trait-based prompts (e.g., Smart, Honest, Criminal, Violent) to investigate potential stereotypical associations; and (3) Physical Characteristics to Country retrieval, focusing on visual attributes like skinny, young, obese, and old to explore how physical appearances are culturally linked to nations. Our findings reveal persistent biases in VLMs, highlighting how visual representations may inadvertently reinforce societal stereotypes.

Gabriela Ben Melech Stan, Estelle Aflalo, Avinash Madasu et al.

Training models on synthetic data has emerged as an increasingly important strategy for improving the performance of generative AI. This approach is particularly helpful for large multimodal models (LMMs) due to the relative scarcity of high-quality paired image-text data compared to language-only data. While a variety of methods have been proposed for generating large multimodal datasets, they do not tailor the synthetic data to address specific deficiencies in the reasoning abilities of LMMs which will be trained with the generated dataset. In contrast, humans often learn in a more efficient manner by seeking out examples related to the types of reasoning where they have failed previously. Inspired by this observation, we propose a new approach for synthetic data generation which is grounded in the analysis of an existing LMM's reasoning failures. Our methodology leverages frontier models to automatically analyze errors produced by a weaker LMM and propose new examples which can be used to correct the reasoning failure via additional training, which are then further filtered to ensure high quality. We generate a large multimodal instruction tuning dataset containing over 553k examples using our approach and conduct extensive experiments demonstrating its utility for improving the performance of LMMs on multiple downstream tasks. Our results show that models trained on our synthetic data can even exceed the performance of LMMs trained on an equivalent amount of additional real data, demonstrating the high value of generating synthetic data targeted to specific reasoning failure modes in LMMs. We will make our dataset and code publicly available.

Avinash Madasu, Vasudev Lal, Phillip Howard

CLIP is one of the most popular foundation models and is heavily used for many vision-language tasks, yet little is known about its inner workings. As CLIP is increasingly deployed in real-world applications, it is becoming even more critical to understand its limitations and embedded social biases to mitigate potentially harmful downstream consequences. However, the question of what internal mechanisms drive both the impressive capabilities as well as problematic shortcomings of CLIP has largely remained unanswered. To bridge this gap, we study the conceptual consistency of text descriptions for attention heads in CLIP-like models. Specifically, we propose Concept Consistency Score (CCS), a novel interpretability metric that measures how consistently individual attention heads in CLIP models align with specific concepts. Our soft-pruning experiments reveal that high CCS heads are critical for preserving model performance, as pruning them leads to a significantly larger performance drop than pruning random or low CCS heads. Notably, we find that high CCS heads capture essential concepts and play a key role in out-of-domain detection, concept-specific reasoning, and video-language understanding. Moreover, we prove that high CCS heads learn spurious correlations which amplify social biases. These results position CCS as a powerful interpretability metric exposing the paradox of performance and social biases in CLIP models.

Avinash Madasu, Vijjini Anvesh Rao

Elastic Weight Consolidation (EWC) is a technique used in overcoming catastrophic forgetting between successive tasks trained on a neural network. We use this phenomenon of information sharing between tasks for domain adaptation. Training data for tasks such as sentiment analysis (SA) may not be fairly represented across multiple domains. Domain Adaptation (DA) aims to build algorithms that leverage information from source domains to facilitate performance on an unseen target domain. We propose a model-independent framework - Sequential Domain Adaptation (SDA). SDA draws on EWC for training on successive source domains to move towards a general domain solution, thereby solving the problem of domain adaptation. We test SDA on convolutional, recurrent, and attention-based architectures. Our experiments show that the proposed framework enables simple architectures such as CNNs to outperform complex state-of-the-art models in domain adaptation of SA. In addition, we observe that the effectiveness of a harder first Anti-Curriculum ordering of source domains leads to maximum performance.

Avinash Madasu, Vijjini Anvesh Rao

Aspect Based Sentiment Analysis (ABSA) is the task of identifying sentiment polarity of a text given another text segment or aspect. In ABSA, a text can have multiple sentiments depending upon each aspect. Aspect Term Sentiment Analysis (ATSA) is a subtask of ABSA, in which aspect terms are contained within the given sentence. Most of the existing approaches proposed for ATSA, incorporate aspect information through a different subnetwork thereby overlooking the advantage of aspect terms' presence within the sentence. In this paper, we propose a model that leverages the positional information of the aspect. The proposed model introduces a decay mechanism based on position. This decay function mandates the contribution of input words for ABSA. The contribution of a word declines as farther it is positioned from the aspect terms in the sentence. The performance is measured on two standard datasets from SemEval 2014 Task 4. In comparison with recent architectures, the effectiveness of the proposed model is demonstrated.

Avinash Madasu, Sivasankar E

Sentiment analysis is a domain of study that focuses on identifying and classifying the ideas expressed in the form of text into positive, negative and neutral polarities. Feature selection is a crucial process in machine learning. In this paper, we aim to study the performance of different feature selection techniques for sentiment analysis. Term Frequency Inverse Document Frequency (TF-IDF) is used as the feature extraction technique for creating feature vocabulary. Various Feature Selection (FS) techniques are experimented to select the best set of features from feature vocabulary. The selected features are trained using different machine learning classifiers Logistic Regression (LR), Support Vector Machines (SVM), Decision Tree (DT) and Naive Bayes (NB). Ensemble techniques Bagging and Random Subspace are applied on classifiers to enhance the performance on sentiment analysis. We show that, when the best FS techniques are trained using ensemble methods achieve remarkable results on sentiment analysis. We also compare the performance of FS methods trained using Bagging, Random Subspace with varied neural network architectures. We show that FS techniques trained using ensemble classifiers outperform neural networks requiring significantly less training time and parameters thereby eliminating the need for extensive hyper-parameter tuning.

Avinash Madasu, Vijjini Anvesh Rao

Text classification has been one of the major problems in natural language processing. With the advent of deep learning, convolutional neural network (CNN) has been a popular solution to this task. However, CNNs which were first proposed for images, face many crucial challenges in the context of text processing, namely in their elementary blocks: convolution filters and max pooling. These challenges have largely been overlooked by the most existing CNN models proposed for text classification. In this paper, we present an experimental study on the fundamental blocks of CNNs in text categorization. Based on this critique, we propose Sequential Convolutional Attentive Recurrent Network (SCARN). The proposed SCARN model utilizes both the advantages of recurrent and convolutional structures efficiently in comparison to previously proposed recurrent convolutional models. We test our model on different text classification datasets across tasks like sentiment analysis and question classification. Extensive experiments establish that SCARN outperforms other recurrent convolutional architectures with significantly less parameters. Furthermore, SCARN achieves better performance compared to equally large various deep CNN and LSTM architectures.

Avinash Madasu, Sivasankar E

Sentiment Analysis refers to the study of systematically extracting the meaning of subjective text . When analysing sentiments from the subjective text using Machine Learning techniques,feature extraction becomes a significant part. We perform a study on the performance of feature extraction techniques TF-IDF(Term Frequency-Inverse Document Frequency) and Doc2vec (Document to Vector) using Cornell movie review datasets, UCI sentiment labeled datasets, stanford movie review datasets,effectively classifying the text into positive and negative polarities by using various pre-processing methods like eliminating StopWords and Tokenization which increases the performance of sentiment analysis in terms of accuracy and time taken by the classifier.The features obtained after applying feature extraction techniques on the text sentences are trained and tested using the classifiers Logistic Regression,Support Vector Machines,K-Nearest Neighbours , Decision Tree and Bernoulli Nave Bayes

Avinash Madasu, Vijjini Anvesh Rao

Domain Adaptation explores the idea of how to maximize performance on a target domain, distinct from source domain, upon which the classifier was trained. This idea has been explored for the task of sentiment analysis extensively. The training of reviews pertaining to one domain and evaluation on another domain is widely studied for modeling a domain independent algorithm. This further helps in understanding correlation between domains. In this paper, we show that Gated Convolutional Neural Networks (GCN) perform effectively at learning sentiment analysis in a manner where domain dependant knowledge is filtered out using its gates. We perform our experiments on multiple gate architectures: Gated Tanh ReLU Unit (GTRU), Gated Tanh Unit (GTU) and Gated Linear Unit (GLU). Extensive experimentation on two standard datasets relevant to the task, reveal that training with Gated Convolutional Neural Networks give significantly better performance on target domains than regular convolution and recurrent based architectures. While complex architectures like attention, filter domain specific knowledge as well, their complexity order is remarkably high as compared to gated architectures. GCNs rely on convolution hence gaining an upper hand through parallelization.

Avinash Madasu, Vijjini Anvesh Rao

Self Normalizing Neural Networks(SNN) proposed on Feed Forward Neural Networks(FNN) outperform regular FNN architectures in various machine learning tasks. Particularly in the domain of Computer Vision, the activation function Scaled Exponential Linear Units (SELU) proposed for SNNs, perform better than other non linear activations such as ReLU. The goal of SNN is to produce a normalized output for a normalized input. Established neural network architectures like feed forward networks and Convolutional Neural Networks(CNN) lack the intrinsic nature of normalizing outputs. Hence, requiring additional layers such as Batch Normalization. Despite the success of SNNs, their characteristic features on other network architectures like CNN haven't been explored, especially in the domain of Natural Language Processing. In this paper we aim to show the effectiveness of proposed, Self Normalizing Convolutional Neural Networks(SCNN) on text classification. We analyze their performance with the standard CNN architecture used on several text classification datasets. Our experiments demonstrate that SCNN achieves comparable results to standard CNN model with significantly fewer parameters. Furthermore it also outperforms CNN with equal number of parameters.