Papers by Megh Thakkar

Chart-to-Text: A Large-Scale Benchmark for Chart Summarization (2022.acl-long)

Copied to clipboard

Shankar Kantharaj, Rixie Tiffany Leong, Xiang Lin, Ahmed Masry, Megh Thakkar, Enamul Hoque, Shafiq Joty

Challenge:	Inferring key insights from charts can be challenging and time-consuming.
Approach:	They propose a task where the goal is to explain a chart and summarize key takeaways from it in natural language.
Outcome:	The proposed model produces fluent summaries but suffers from hallucinations and factual errors . the proposed model is compared with other models and can be used to generate BLEU scores .

CIAug: Equipping Interpolative Augmentation with Curriculum Learning (2022.naacl-main)

Copied to clipboard

Ramit Sawhney, Ritesh Soun, Shrey Pandit, Megh Thakkar, Sarvagya Malaviya, Yuval Pinter

Language Models

Challenge:	Current methods for interpolative data augmentation select samples at random, which might make it difficult for the model to generalize better and converge faster.
Approach:	They propose a curriculum-based learning method that leverages the relative position of samples in hyperbolic embedding space as a complexity measure to gradually mix up increasingly difficult and diverse samples along training.
Outcome:	The proposed method achieves state-of-the-art results over existing methods on 10 benchmark datasets across 4 languages in text classification and named-entity recognition tasks.

Self-Influence Guided Data Reweighting for Language Model Pre-training (2023.emnlp-main)

Copied to clipboard

Megh Thakkar, Tolga Bolukbasi, Sriram Ganapathy, Shikhar Vashishth, Sarath Chandar, Partha Talukdar

Language Models

Challenge:	Language Models (LMs) pre-trained with selfsupervision on large text data are the default starting point for developing models for various downstream tasks.
Approach:	They propose a method for jointly reweighting samples by leveraging self-influence scores as an indicator of sample importance and pre-training.
Outcome:	The proposed method promotes novelty and stability for model pre-training.

Towards Robust Low-Resource Fine-Tuning with Multi-View Compressed Representations (2023.acl-long)

Copied to clipboard

Linlin Liu, Xingxuan Li, Megh Thakkar, Xin Li, Shafiq Joty, Luo Si, Lidong Bing

Responsible & Trustworthy NLP

Challenge:	Using hidden representations, pretrained language models are prone to overfitting due to the huge amount of parameters.
Approach:	They propose a method that inserts random autoencoders between hidden layers of a PLM to transform activations from the previous layers into multi-view compressed representations before feeding them into the upper layers.
Outcome:	The proposed method improves performance across sequence- and token-level lowresource tasks.

ColMate: Contrastive Late Interaction and Masked Text for Multimodal Document Retrieval (2025.emnlp-industry)

Copied to clipboard

Ahmed Masry, Megh Thakkar, Patrice Bechard, Sathwik Tejaswi Madhusudhan, Rabiul Awal, Shambhavi Mishra, Akshay Kalkunte Suresh, Srivatsava Daruru, Enamul Hoque, Spandana Gella, Torsten Scholak, Sai Rajeswar

Information Retrieval

Challenge:	Existing methods for multimodal document retrieval often replicate techniques developed for text-only retrieval.
Approach:	They propose a document retrieval model that bridges the gap between multimodal representation learning and document retrievals by providing external knowledge as context.
Outcome:	The proposed model achieves 3.61% improvement over existing retrieval models on the ViDoRe V2 benchmark, showing stronger generalization to out-of-domain benchmarks.

Randomized Smoothing with Masked Inference for Adversarially Robust Text Classifications (2023.acl-long)

Copied to clipboard

Han Cheol Moon, Shafiq Joty, Ruochen Zhao, Megh Thakkar, Chi Xu

Robustness in NLP

Challenge:	Large-scale pre-trained language models are brittle against specifically crafted adversarial examples, leading to increasing interest in probing the adversariality of NLP systems.
Approach:	They propose a two-stage framework that combines randomized smoothing and masked inference to improve the adversarial robustness of NLP systems.
Outcome:	The proposed framework improves adversarial robustness by 2 to 3 times over existing state-of-the-art methods on benchmark datasets.

AdaPT: A Set of Guidelines for Hyperbolic Multimodal Multilingual NLP (2024.findings-naacl)

Copied to clipboard

Ramit Sawhney, Shrey Pandit, Vishwa Shah, Megh Thakkar, Shafiq Joty

Multimodality

Challenge:	Euclidean space is used for training neural models and performing arithmetic operations, but many data types have complex geometries and cannot be captured in the Euclidesan space.
Approach:	They propose a set of guidelines for initialization, parametrization, and training of neural networks that can be generalized over existing neural network training methodologies.
Outcome:	The proposed framework outperforms Euclidean methods on three tasks over 12 languages and modalities on a variety of domains.

Combining Domain and Alignment Vectors Provides Better Knowledge-Safety Trade-offs in LLMs (2025.acl-short)

Copied to clipboard

Megh Thakkar, Quentin Fournier, Matthew Riemer, Pin-Yu Chen, Amal Zouaq, Payel Das, Sarath Chandar

Responsible & Trustworthy NLP

Challenge:	Large language models (LLMs) excel in specific technical fields, but are not explicitly trained to be safe.
Approach:	They propose a model merging-based alignment method that allows for safer domain-specific models that preserve their utility.
Outcome:	The proposed method improves safety alignment on LLMs with minimal degradation on domain-specific benchmarks.

Tweet Based Reach Aware Temporal Attention Network for NFT Valuation (2022.findings-emnlp)

Copied to clipboard

Ramit Sawhney, Megh Thakkar, Ritesh Soun, Atula Neerkaje, Vasu Sharma, Dipanwita Guhathakurta, Sudheer Chava

Sentiment Analysis

Challenge:	Non-Fungible Tokens (NFTs) are a relatively unexplored class of assets due to their extremely volatile nature.
Approach:	They propose a reach-aware temporal learning approach to predict future NFT trends from a dataset consisting of over 1.3 million tweets and 180 thousand NFT transactions .
Outcome:	The proposed model outperforms state-of-the-art models by an average of 36% on a dataset consisting of over 1.3 million tweets and 180 thousand NFT transactions spanning over 15 NFT collections.

ChartQAPro: A More Diverse and Challenging Benchmark for Chart Question Answering (2025.findings-acl)

Copied to clipboard

Ahmed Masry, Mohammed Saidul Islam, Mahir Ahmed, Aayush Bajaj, Firoz Kabir, Aaryaman Kartha, Md Tahmid Rahman Laskar, Mizanur Rahman, Shadikur Rahman, Mehrad Shahmohammadi, Megh Thakkar, Md Rizwan Parvez, Enamul Hoque, Shafiq Joty

Natural Language Interfaces

Challenge:	Chart Question Answering systems are limited in their ability to interpret data visually and reason with visual representations.
Approach:	They propose a chart-based chart question-answering system that includes 1,341 charts from 99 diverse sources and 1,948 questions in various types.
Outcome:	The new benchmark includes 1,341 charts from 99 diverse sources and 1,948 questions in various types.

A Deep Dive into the Trade-Offs of Parameter-Efficient Preference Alignment Techniques (2024.acl-long)

Copied to clipboard

Megh Thakkar, Quentin Fournier, Matthew Riemer, Pin-Yu Chen, Amal Zouaq, Payel Das, Sarath Chandar

Green & Sustainable NLP

Challenge:	Large language models are pre-trained on trillions of tokens and instruction-tuned or aligned to specific preferences.
Approach:	They propose guidelines to help researchers perform more effective parameter-efficient LLM alignment.
Outcome:	The proposed methods outperform preference optimization and outperformed pre-trained models on three key axes.

HypMix: Hyperbolic Interpolative Data Augmentation (2021.emnlp-main)

Copied to clipboard

Ramit Sawhney, Megh Thakkar, Shivam Agarwal, Di Jin, Diyi Yang, Lucie Flek

Responsible & Trustworthy NLP

Challenge:	Existing methods for data augmentation involve performing mathematical operations over the raw input samples or their latent states representations, but these operations are performed in the Euclidean space, simplifying these representations and resulting in noisy interpolations.
Approach:	They propose a model-, data-, and modality-agnostic interpolative data augmentation technique operating in the hyperbolic space that captures the complex geometry of input and hidden state hierarchies better than its contemporaries.
Outcome:	The proposed technique outperforms state-of-the-art methods on benchmark and low resource datasets across speech, text, and vision modalities.

ChartGemma: Visual Instruction-tuning for Chart Reasoning in the Wild (2025.coling-industry)

Copied to clipboard

Ahmed Masry, Megh Thakkar, Aayush Bajaj, Aaryaman Kartha, Enamul Hoque, Shafiq Joty

Multimodality

Challenge:	Existing methods for chart understanding and reasoning are weakly aligned and rely on underlying data tables.
Approach:	They propose a chart-based understanding and reasoning model that is trained on instruction-tuning data generated directly from chart images.
Outcome:	The proposed model achieves state-of-the-art results across 5 benchmarks spanning chart summarization, question answering, and fact-checking.

DMix: Adaptive Distance-aware Interpolative Mixup (2022.acl-short)

Copied to clipboard

Ramit Sawhney, Megh Thakkar, Shrey Pandit, Ritesh Soun, Di Jin, Diyi Yang, Lucie Flek

Challenge:	Interpolation-based regularisation methods such as Mixup have shown to be effective for various tasks and modalities.
Approach:	They propose an adaptive distance-aware interpolative Mixup that selects samples based on their diversity in the embedding space.
Outcome:	The proposed method achieves state-of-the-art on sentence classification over existing methods on 8 benchmark datasets across English, Arabic, Turkish, and Hindi languages while achieving benchmark F1 scores in 3 times less number of iterations.