Search

Having a clean dataset has been the foundational assumption of most natural language processing (NLP) systems. However, properly written text is rarely found in real-world scenarios and hence, oftentimes invalidates the aforementioned foundational assumption. Recently, Large language models (LLMs) have shown impressive performance, but can they handle the inevitable noise in real-world data? This work tackles this critical question by investigating LLMs' resilience against morphological variations in text. To that end, we artificially introduce varying levels of noise into a diverse set of datasets and systematically evaluate LLMs' robustness against the corrupt variations of the original text. Our findings show that contrary to popular beliefs, generative LLMs are quiet robust to noisy perturbations in text. This is a departure from pre-trained models like BERT or RoBERTa whose performance has been shown to be sensitive to deteriorating noisy text. Additionally, we test LLMs' resilience on multiple real-world benchmarks that closely mimic commonly found errors in the wild. With minimal prompting, LLMs achieve a new state-of-the-art on the benchmark tasks of Grammar Error Correction (GEC) and Lexical Semantic Change (LSC). To empower future research, we also release a dataset annotated by humans stating their preference for LLM vs. human-corrected outputs along with the code to reproduce our results., Comment: 8 pages, 1 figure, 6 tables, updated with results also from GPT-4, LLaMa-3

Large language models (LLMs) have shown remarkable performance on many tasks in different domains. However, their performance in closed-book biomedical machine reading comprehension (MRC) has not been evaluated in depth. In this work, we evaluate GPT on four closed-book biomedical MRC benchmarks. We experiment with different conventional prompting techniques as well as introduce our own novel prompting method. To solve some of the retrieval problems inherent to LLMs, we propose a prompting strategy named Implicit Retrieval Augmented Generation (RAG) that alleviates the need for using vector databases to retrieve important chunks in traditional RAG setups. Moreover, we report qualitative assessments on the natural language generation outputs from our approach. The results show that our new prompting technique is able to get the best performance in two out of four datasets and ranks second in rest of them. Experiments show that modern-day LLMs like GPT even in a zero-shot setting can outperform supervised models, leading to new state-of-the-art (SoTA) results on two of the benchmarks.

The mining of adverse drug events (ADEs) is pivotal in pharmacovigilance, enhancing patient safety by identifying potential risks associated with medications, facilitating early detection of adverse events, and guiding regulatory decision-making. Traditional ADE detection methods are reliable but slow, not easily adaptable to large-scale operations, and offer limited information. With the exponential increase in data sources like social media content, biomedical literature, and Electronic Medical Records (EMR), extracting relevant ADE-related information from these unstructured texts is imperative. Previous ADE mining studies have focused on text-based methodologies, overlooking visual cues, limiting contextual comprehension, and hindering accurate interpretation. To address this gap, we present a MultiModal Adverse Drug Event (MMADE) detection dataset, merging ADE-related textual information with visual aids. Additionally, we introduce a framework that leverages the capabilities of LLMs and VLMs for ADE detection by generating detailed descriptions of medical images depicting ADEs, aiding healthcare professionals in visually identifying adverse events. Using our MMADE dataset, we showcase the significance of integrating visual cues from images to enhance overall performance. This approach holds promise for patient safety, ADE awareness, and healthcare accessibility, paving the way for further exploration in personalized healthcare., Comment: ACL Findings 2024

Electronic health records (EHR) even though a boon for healthcare practitioners, are growing convoluted and longer every day. Sifting around these lengthy EHRs is taxing and becomes a cumbersome part of physician-patient interaction. Several approaches have been proposed to help alleviate this prevalent issue either via summarization or sectioning, however, only a few approaches have truly been helpful in the past. With the rise of automated methods, machine learning (ML) has shown promise in solving the task of identifying relevant sections in EHR. However, most ML methods rely on labeled data which is difficult to get in healthcare. Large language models (LLMs) on the other hand, have performed impressive feats in natural language processing (NLP), that too in a zero-shot manner, i.e. without any labeled data. To that end, we propose using LLMs to identify relevant section headers. We find that GPT-4 can effectively solve the task on both zero and few-shot settings as well as segment dramatically better than state-of-the-art methods. Additionally, we also annotate a much harder real world dataset and find that GPT-4 struggles to perform well, alluding to further research and harder benchmarks., Comment: To appear in NAACL 2024 at the 6th Clinical Natural Language Processing Workshop

In this paper, our focus lies on the Merton's jump diffusion model, employing jump processes characterized by the compound Poisson process. Our primary objective is to forecast the drift and volatility of the model using a variety of methodologies. We adopt an approach that involves implementing different drift, volatility, and jump terms within the model through various machine learning techniques, traditional methods, and statistical methods on price-volume data. Additionally, we introduce a path-dependent Monte Carlo simulation to model cryptocurrency prices, taking into account the volatility and unexpected jumps in prices., Comment: 21 pages

Health insurance companies have a defined process called prior authorization (PA) which is a health plan cost-control process that requires doctors and other healthcare professionals to get clearance in advance from a health plan before performing a particular procedure on a patient in order to be eligible for payment coverage. For health insurance companies, approving PA requests for patients in the medical domain is a time-consuming and challenging task. One of those key challenges is validating if a request matches up to certain criteria such as age, gender, etc. In this work, we evaluate whether GPT can validate numerous key factors, in turn helping health plans reach a decision drastically faster. We frame it as a question answering task, prompting GPT to answer a question from patient electronic health record. We experiment with different conventional prompting techniques as well as introduce our own novel prompting technique. Moreover, we report qualitative assessment by humans on the natural language generation outputs from our approach. Results show that our method achieves superior performance with the mean weighted F1 score of 0.61 as compared to its standard counterparts.

Prompt engineering has emerged as an indispensable technique for extending the capabilities of large language models (LLMs) and vision-language models (VLMs). This approach leverages task-specific instructions, known as prompts, to enhance model efficacy without modifying the core model parameters. Rather than updating the model parameters, prompts allow seamless integration of pre-trained models into downstream tasks by eliciting desired model behaviors solely based on the given prompt. Prompts can be natural language instructions that provide context to guide the model or learned vector representations that activate relevant knowledge. This burgeoning field has enabled success across various applications, from question-answering to commonsense reasoning. However, there remains a lack of systematic organization and understanding of the diverse prompt engineering methods and techniques. This survey paper addresses the gap by providing a structured overview of recent advancements in prompt engineering, categorized by application area. For each prompting approach, we provide a summary detailing the prompting methodology, its applications, the models involved, and the datasets utilized. We also delve into the strengths and limitations of each approach and include a taxonomy diagram and table summarizing datasets, models, and critical points of each prompting technique. This systematic analysis enables a better understanding of this rapidly developing field and facilitates future research by illuminating open challenges and opportunities for prompt engineering., Comment: 9 pages, 2 figures

In this work, we focus on label efficient learning for video action detection. We develop a novel semi-supervised active learning approach which utilizes both labeled as well as unlabeled data along with informative sample selection for action detection. Video action detection requires spatio-temporal localization along with classification, which poses several challenges for both active learning informative sample selection as well as semi-supervised learning pseudo label generation. First, we propose NoiseAug, a simple augmentation strategy which effectively selects informative samples for video action detection. Next, we propose fft-attention, a novel technique based on high-pass filtering which enables effective utilization of pseudo label for SSL in video action detection by emphasizing on relevant activity region within a video. We evaluate the proposed approach on three different benchmark datasets, UCF-101-24, JHMDB-21, and Youtube-VOS. First, we demonstrate its effectiveness on video action detection where the proposed approach outperforms prior works in semi-supervised and weakly-supervised learning along with several baseline approaches in both UCF101-24 and JHMDB-21. Next, we also show its effectiveness on Youtube-VOS for video object segmentation demonstrating its generalization capability for other dense prediction tasks in videos. The code and models is publicly available at: \url{https://github.com/AKASH2907/semi-sup-active-learning}., Comment: AAAI Conference on Artificial Intelligence, Main Technical Track (AAAI), 2024, Code: https://github.com/AKASH2907/semi-sup-active-learning

Global contexts in images are quite valuable in image-to-image translation problems. Conventional attention-based and graph-based models capture the global context to a large extent, however, these are computationally expensive. Moreover, the existing approaches are limited to only learning the pairwise semantic relation between any two points on the image. In this paper, we present Latent Graph Attention (LGA) a computationally inexpensive (linear to the number of nodes) and stable, modular framework for incorporating the global context in the existing architectures, especially empowering small-scale architectures to give performance closer to large size architectures, thus making the light-weight architectures more useful for edge devices with lower compute power and lower energy needs. LGA propagates information spatially using a network of locally connected graphs, thereby facilitating to construct a semantically coherent relation between any two spatially distant points that also takes into account the influence of the intermediate pixels. Moreover, the depth of the graph network can be used to adapt the extent of contextual spread to the target dataset, thereby being able to explicitly control the added computational cost. To enhance the learning mechanism of LGA, we also introduce a novel contrastive loss term that helps our LGA module to couple well with the original architecture at the expense of minimal additional computational load. We show that incorporating LGA improves the performance on three challenging applications, namely transparent object segmentation, image restoration for dehazing and optical flow estimation., Comment: 20 pages, 7 figures

We demonstrate a versatile technique for generating continuously wavelength-tunable laser waveforms, with mJ pulse energies and ultrashort pulse durations down to few-cycle in the ultraviolet C and visible spectral ranges. Using the processes of self-phase modulation or Raman-induced spectral broadening, we substantially expand the spectrum of a femtosecond 1030 nm Yb:CaF$_{2}$ laser, allowing for an extensive wavelength-tunability of the second and fourth harmonics of the laser within the visible and ultraviolet C spectral regions at 460-580 nm and 230-290 nm. In addition, our approach exploits nonlinearly assisted self-compression in the second harmonic upconversion of the spectrally broadened infrared pulses with high-order dispersion. This results in 8-30 femtosecond pulses in the visible with an intensity enhancement of up to 30 times. Such an ultrashort visible and ultraviolet C source is ideal for investigating ultrafast dynamics in molecules, solids, and bio and nano systems. Furthermore, this tunable light source enables the generation of bright, narrow-bandwidth, continuously wavelength-tunable, coherent light in the extreme ultraviolet to soft X-ray spectral region. Theoretically, the X-ray pulse structure can consist of sub-200 as pulse trains, making such a source highly suitable for dynamic multidimensional imaging. This includes coherent diffractive imaging of ferromagnetic nanostructures where resonant X-ray scattering is essential, as well as X-ray absorption spectroscopies of advanced quantum materials at pico-nanometer spatial and atto-femtosecond temporal scales.

Gas leakage is a serious hazard in a variety of sectors, including industrial, domestic, and gas-powered vehicles. The installation of gas leak detection systems has emerged as a critical preventative measure to solve this problem. Traditional gas sensors, such as electrochemical, infrared point, and Metal Oxide Semiconductor sensors, have been widely used for gas leak detection. However, these sensors have limitations in terms of their adaptation to various gases, as well as their high cost and difficulties in scaling. In this paper, a novel non-contact gas detection technique based on a 40 kHz ultrasonic signal is described. The proposed approach employs the reflections of the emitted ultrasonic wave to detect the gas leaks and also able to identify the exact gas in real-time. To confirm the method's effectiveness, trials were carried out using Hydrogen, Helium, Argon, and Butane gas. The system identified gas flow breaches in 0.01 seconds, whereas the gas identification procedure took 0.8 seconds. An interesting extension of the proposed approach is real-time visualisation of gas flow employing an array of transducers.

The creation of unique control methods for a hand prosthesis is still a problem that has to be addressed. The best choice of a human-machine interface (HMI) that should be used to enable natural control is still a challenge. Surface electromyography (sEMG), the most popular option, has a variety of difficult-to-fix issues (electrode displacement, sweat, fatigue). The ultrasound imaging-based methodology offers a means of recognising complex muscle activity and configuration with a greater SNR and less hardware requirements as compared to sEMG. In this study, a prototype system for high frame rate ultrasound imaging for prosthetic arm control is proposed. Using the proposed framework, a virtual robotic hand simulation is developed that can mimic a human hand as illustrated in the link [10]. The proposed classification model simulating four hand gestures has a classification accuracy of more than 90%.

State-of-the-art pre-trained language models (PLMs) outperform other models when applied to the majority of language processing tasks. However, PLMs have been found to degrade in performance under distribution shift, a phenomenon that occurs when data at test-time does not come from the same distribution as the source training set. Equally as challenging is the task of obtaining labels in real-time due to issues like long-labeling feedback loops. The lack of adequate methods that address the aforementioned challenges constitutes the need for approaches that continuously adapt the PLM to a distinct distribution. Unsupervised domain adaptation adapts a source model to an unseen as well as unlabeled target domain. While some techniques such as data augmentation can adapt models in several scenarios, they have only been sparsely studied for addressing the distribution shift problem. In this work, we present an approach (MEMO-CL) that improves the performance of PLMs at test-time under distribution shift. Our approach takes advantage of the latest unsupervised techniques in data augmentation and adaptation to minimize the entropy of the PLM's output distribution. MEMO-CL operates on a batch of augmented samples from a single observation in the test set. The technique introduced is unsupervised, domain-agnostic, easy to implement, and requires no additional data. Our experiments result in a 3% improvement over current test-time adaptation baselines., Comment: Accepted to 2nd International Workshop on Practical Deep Learning in the Wild at AAAI 2023

Large language models (LLMs) have been shown to be able to perform new tasks based on a few demonstrations or natural language instructions. While these capabilities have led to widespread adoption, most LLMs are developed by resource-rich organizations and are frequently kept from the public. As a step towards democratizing this powerful technology, we present BLOOM, a 176B-parameter open-access language model designed and built thanks to a collaboration of hundreds of researchers. BLOOM is a decoder-only Transformer language model that was trained on the ROOTS corpus, a dataset comprising hundreds of sources in 46 natural and 13 programming languages (59 in total). We find that BLOOM achieves competitive performance on a wide variety of benchmarks, with stronger results after undergoing multitask prompted finetuning. To facilitate future research and applications using LLMs, we publicly release our models and code under the Responsible AI License.

Training and evaluating language models increasingly requires the construction of meta-datasets --diverse collections of curated data with clear provenance. Natural language prompting has recently lead to improved zero-shot generalization by transforming existing, supervised datasets into a diversity of novel pretraining tasks, highlighting the benefits of meta-dataset curation. While successful in general-domain text, translating these data-centric approaches to biomedical language modeling remains challenging, as labeled biomedical datasets are significantly underrepresented in popular data hubs. To address this challenge, we introduce BigBIO a community library of 126+ biomedical NLP datasets, currently covering 12 task categories and 10+ languages. BigBIO facilitates reproducible meta-dataset curation via programmatic access to datasets and their metadata, and is compatible with current platforms for prompt engineering and end-to-end few/zero shot language model evaluation. We discuss our process for task schema harmonization, data auditing, contribution guidelines, and outline two illustrative use cases: zero-shot evaluation of biomedical prompts and large-scale, multi-task learning. BigBIO is an ongoing community effort and is available at https://github.com/bigscience-workshop/biomedical, Comment: Submitted to NeurIPS 2022 Datasets and Benchmarks Track

Natural interface to database (NLIDB) has been researched a lot during the past decades. In the core of NLIDB, is a semantic parser used to convert natural language into SQL. Solutions from traditional NLP methodology focuses on grammar rule pattern learning and pairing via intermediate logic forms. Although those methods give an acceptable performance on certain specific database and parsing tasks, they are hard to generalize and scale. On the other hand, recent progress in neural deep learning seems to provide a promising direction towards building a general NLIDB system. Unlike the traditional approach, those neural methodologies treat the parsing problem as a sequence-to-sequence learning problem. In this paper, we experimented on several sequence-to-sequence learning models and evaluate their performance on general database parsing task., Comment: Semester project report for NLP course

At a discrete set of magic angles, twisted bilayer graphene has been shown to host extraordinarily flat bands, correlated insulating states, unconventional superconductivity, and distinct Landau level degeneracies. In this work, we design a highly efficient quantum Otto engine using a twisted bilayer graphene sample. Flat bands, which occur at magic angles, make the prospect of extracting useful work from our Otto engine lucrative. We use an eight-band continuum model of twisted bilayer graphene to compute efficiencies and work outputs for magic and non-magic angle twists, and compare the results with an $AB$ stacked bilayer and a monolayer. It is observed that the efficiency varies smoothly with the twist angle and the maximum is attained at the magic angle., Comment: 9 pages, 8 figures, accepted for publication in Physical Review B

Learning to dehaze single hazy images, especially using a small training dataset is quite challenging. We propose a novel generative adversarial network architecture for this problem, namely back projected pyramid network (BPPNet), that gives good performance for a variety of challenging haze conditions, including dense haze and inhomogeneous haze. Our architecture incorporates learning of multiple levels of complexities while retaining spatial context through iterative blocks of UNets and structural information of multiple scales through a novel pyramidal convolution block. These blocks together for the generator and are amenable to learning through back projection. We have shown that our network can be trained without over-fitting using as few as 20 image pairs of hazy and non-hazy images. We report the state of the art performances on NTIRE 2018 homogeneous haze datasets for indoor and outdoor images, NTIRE 2019 denseHaze dataset, and NTIRE 2020 non-homogeneous haze dataset., Comment: 16 pages, 8 figures, to be published in Computer Vision ECCV 2020 Workshops

Fetal magnetic resonance imaging (MRI) is challenged by uncontrollable, large, and irregular fetal movements. It is, therefore, performed through visual monitoring of fetal motion and repeated acquisitions to ensure diagnostic-quality images are acquired. Nevertheless, visual monitoring of fetal motion based on displayed slices, and navigation at the level of stacks-of-slices is inefficient. The current process is highly operator-dependent, increases scanner usage and cost, and significantly increases the length of fetal MRI scans which makes them hard to tolerate for pregnant women. To help build automatic MRI motion tracking and navigation systems to overcome the limitations of the current process and improve fetal imaging, we have developed a new real time image-based motion tracking method based on deep learning that learns to predict fetal motion directly from acquired images. Our method is based on a recurrent neural network, composed of spatial and temporal encoder-decoders, that infers motion parameters from anatomical features extracted from sequences of acquired slices. We compared our trained network on held out test sets (including data with different characteristics, e.g. different fetuses scanned at different ages, and motion trajectories recorded from volunteer subjects) with networks designed for estimation as well as methods adopted to make predictions. The results show that our method outperformed alternative techniques, and achieved real-time performance with average errors of 3.5 and 8 degrees for the estimation and prediction tasks, respectively. Our real-time deep predictive motion tracking technique can be used to assess fetal movements, to guide slice acquisitions, and to build navigation systems for fetal MRI., Comment: The article has been published in IEEE TMI: 14 pages, 11 figures, 2 tables and 1 supplementary https://github.com/bchimagine/DeepPredictiveMotionTracking

Expressing in language is subjective. Everyone has a different style of reading and writing, apparently it all boil downs to the way their mind understands things (in a specific format). Language style transfer is a way to preserve the meaning of a text and change the way it is expressed. Progress in language style transfer is lagged behind other domains, such as computer vision, mainly because of the lack of parallel data, use cases, and reliable evaluation metrics. In response to the challenge of lacking parallel data, we explore learning style transfer from non-parallel data. We propose a model combining seq2seq, autoencoders, and adversarial loss to achieve this goal. The key idea behind the proposed models is to learn separate content representations and style representations using adversarial networks. Considering the problem of evaluating style transfer tasks, we frame the problem as sentiment transfer and evaluation using a sentiment classifier to calculate how many sentiments was the model able to transfer. We report our results on several kinds of models., Comment: Report built as a part of project for CSYE7245 Northeastern University under Prof. Nik Brown. arXiv admin note: text overlap with arXiv:1711.06861, arXiv:1409.3215, arXiv:1705.07663 by other authors