Jay Patel

Shivalika Singh, Freddie Vargus, Daniel Dsouza et al. · cmu

Datasets are foundational to many breakthroughs in modern artificial intelligence. Many recent achievements in the space of natural language processing (NLP) can be attributed to the finetuning of pre-trained models on a diverse set of tasks that enables a large language model (LLM) to respond to instructions. Instruction fine-tuning (IFT) requires specifically constructed and annotated datasets. However, existing datasets are almost all in the English language. In this work, our primary goal is to bridge the language gap by building a human-curated instruction-following dataset spanning 65 languages. We worked with fluent speakers of languages from around the world to collect natural instances of instructions and completions. Furthermore, we create the most extensive multilingual collection to date, comprising 513 million instances through templating and translating existing datasets across 114 languages. In total, we contribute four key resources: we develop and open-source the Aya Annotation Platform, the Aya Dataset, the Aya Collection, and the Aya Evaluation Suite. The Aya initiative also serves as a valuable case study in participatory research, involving collaborators from 119 countries. We see this as a valuable framework for future research collaborations that aim to bridge gaps in resources.

Francesco Picetti, Shrinath Deshpande, Jonathan Leban et al.

We present a novel human body model formulated by an extensive set of anthropocentric measurements, which is capable of generating a wide range of human body shapes and poses. The proposed model enables direct modeling of specific human identities through a deep generative architecture, which can produce humans in any arbitrary pose. It is the first of its kind to have been trained end-to-end using only synthetically generated data, which not only provides highly accurate human mesh representations but also allows for precise anthropometry of the body. Moreover, using a highly diverse animation library, we articulated our synthetic humans' body and hands to maximize the diversity of the learnable priors for model training. Our model was trained on a dataset of $100k$ procedurally-generated posed human meshes and their corresponding anthropometric measurements. Our synthetic data generator can be used to generate millions of unique human identities and poses for non-commercial academic research purposes.

Jay Patel, Joel Chan

Across scholarly communities, manuscripts face similar evaluative rituals: editors invite experts to privately assess submissions through formal peer reviews. This closed, loosely structured, and publisher-mediated process is now being supplemented by critiques on open, distributed platforms. We call this practice, a blend of three open peer review variants, informal peer review as it is accessible to outsiders, unmediated by publishers, and conducted across public platforms. Informal peer reviewers range from occasional error detectors to experienced sleuths who identify plagiarism, fraud, errors, conflicts of interest, and conceptual flaws. They may interpret methods, clarify jargon, assess value, and connect to related work. Here, we asked four questions: (1) Who are informal peer reviewers? (2) Where do they work? (3) How do they evaluate research? and (4) What are their impacts? To answer these questions, we conducted a cross-platform digital ethnography with participant observation. We traced discourse across communities over four months and revisited cases after nine and twelve months. From 15 communities, we selected 12 case mentions (10 unique cases) and 8 meta-commentaries from 26 reviewers. Using open and axial coding, we generated 1,080 codes and four themes: reviewers are a motley crew, they self-organize across subpar digital spaces, use deep, uncommon strategies, and they face resistance from authors, publishers, and editors. Informal peer review, we concluded, is a fragile, minimally governed patchwork of people, platforms, and practices, as well as an emerging evidence infrastructure that can be scaled up. We advise advocates and tool-builders to evolve informal review tools, communities, training, and governance by connecting to scholars' values, reducing participation friction, and rewarding attempts to extend the scholarly dialogue.

Jay Patel, Hrudayangam Mehta, Jeremy Blackburn

Detecting hateful content is a challenging and important problem. Automated tools, like machine-learning models, can help, but they require continuous training to adapt to the ever-changing landscape of social media. In this work, we evaluate eight open-source LLMs' capability to detect antisemitic content, specifically leveraging in-context definition. We also study how LLMs understand and explain their decisions given a moderation policy as a guideline. First, we explore various prompting techniques and design a new CoT-like prompt, Guided-CoT, and find that injecting domain-specific thoughts increases performance and utility. Guided-CoT handles the in-context policy well, improving performance and utility by reducing refusals across all evaluated models, regardless of decoding configuration, model size, or reasoning capability. Notably, Llama 3.1 70B outperforms fine-tuned GPT-3.5. Additionally, we examine LLM errors and introduce metrics to quantify semantic divergence in model-generated rationales, revealing notable differences and paradoxical behaviors among LLMs. Our experiments highlight the differences observed across LLMs' utility, explainability, and reliability. Code and resources available at: https://github.com/idramalab/quantify-llm-explanations

Raghav Mehta, Angelos Filos, Ujjwal Baid et al.

Deep learning (DL) models have provided state-of-the-art performance in various medical imaging benchmarking challenges, including the Brain Tumor Segmentation (BraTS) challenges. However, the task of focal pathology multi-compartment segmentation (e.g., tumor and lesion sub-regions) is particularly challenging, and potential errors hinder translating DL models into clinical workflows. Quantifying the reliability of DL model predictions in the form of uncertainties could enable clinical review of the most uncertain regions, thereby building trust and paving the way toward clinical translation. Several uncertainty estimation methods have recently been introduced for DL medical image segmentation tasks. Developing scores to evaluate and compare the performance of uncertainty measures will assist the end-user in making more informed decisions. In this study, we explore and evaluate a score developed during the BraTS 2019 and BraTS 2020 task on uncertainty quantification (QU-BraTS) and designed to assess and rank uncertainty estimates for brain tumor multi-compartment segmentation. This score (1) rewards uncertainty estimates that produce high confidence in correct assertions and those that assign low confidence levels at incorrect assertions, and (2) penalizes uncertainty measures that lead to a higher percentage of under-confident correct assertions. We further benchmark the segmentation uncertainties generated by 14 independent participating teams of QU-BraTS 2020, all of which also participated in the main BraTS segmentation task. Overall, our findings confirm the importance and complementary value that uncertainty estimates provide to segmentation algorithms, highlighting the need for uncertainty quantification in medical image analyses. Finally, in favor of transparency and reproducibility, our evaluation code is made publicly available at: https://github.com/RagMeh11/QU-BraTS.

Nishanth Arun, Nathan Gaw, Praveer Singh et al.

Saliency maps have become a widely used method to make deep learning models more interpretable by providing post-hoc explanations of classifiers through identification of the most pertinent areas of the input medical image. They are increasingly being used in medical imaging to provide clinically plausible explanations for the decisions the neural network makes. However, the utility and robustness of these visualization maps has not yet been rigorously examined in the context of medical imaging. We posit that trustworthiness in this context requires 1) localization utility, 2) sensitivity to model weight randomization, 3) repeatability, and 4) reproducibility. Using the localization information available in two large public radiology datasets, we quantify the performance of eight commonly used saliency map approaches for the above criteria using area under the precision-recall curves (AUPRC) and structural similarity index (SSIM), comparing their performance to various baseline measures. Using our framework to quantify the trustworthiness of saliency maps, we show that all eight saliency map techniques fail at least one of the criteria and are, in most cases, less trustworthy when compared to the baselines. We suggest that their usage in the high-risk domain of medical imaging warrants additional scrutiny and recommend that detection or segmentation models be used if localization is the desired output of the network. Additionally, to promote reproducibility of our findings, we provide the code we used for all tests performed in this work at this link: https://github.com/QTIM-Lab/Assessing-Saliency-Maps.

Manon A. Dorster, Felix J. Dorfner, Mason C. Cleveland et al.

As early detection of breast cancer strongly favors successful therapeutic outcomes, there is major commercial interest in optimizing breast cancer screening. However, current risk prediction models achieve modest performance and do not incorporate digital breast tomosynthesis (DBT) imaging, which was FDA-approved for breast cancer screening in 2011. To address this unmet need, we present a deep learning (DL)-based framework capable of forecasting an individual patient's 5-year breast cancer risk directly from screening DBT. Using an unparalleled dataset of 161,753 DBT examinations from 50,590 patients, we trained a risk predictor based on features extracted using the Meta AI DINOv2 image encoder, combined with a cumulative hazard layer, to assess a patient's likelihood of developing breast cancer over five years. On a held-out test set, our best-performing model achieved an AUROC of 0.80 on predictions within 5 years. These findings reveal the high potential of DBT-based DL approaches to complement traditional risk assessment tools, and serve as a promising basis for additional investigation to validate and enhance our work.

Eion Tyacke, Kunal Gupta, Jay Patel et al.

Hand gestures are a primary output of the human motor system, yet the decoding of their neuromuscular signatures remains a bottleneck for basic neuroscience and assistive technologies such as prosthetics. Traditional human-machine interface pipelines rely on a single biosignal modality, but multimodal fusion can exploit complementary information from sensors. We systematically compare linear and attention-based fusion strategies across three architectures: a Multimodal MLP, a Multimodal Transformer, and a Hierarchical Transformer, evaluating performance on scenarios with unimodal and multimodal inputs. Experiments use two publicly available datasets: NinaPro DB2 (sEMG and accelerometer) and HD-sEMG 65-Gesture (high-density sEMG and force). Across both datasets, the Hierarchical Transformer with attention-based fusion consistently achieved the highest accuracy, surpassing the multimodal and best single-modality linear-fusion MLP baseline by over 10% on NinaPro DB2 and 3.7% on HD-sEMG. To investigate how modalities interact, we introduce an Isolation Network that selectively silences unimodal or cross-modal attention pathways, quantifying each group of token interactions' contribution to downstream decisions. Ablations reveal that cross-modal interactions contribute approximately 30% of the decision signal across transformer layers, highlighting the importance of attention-driven fusion in harnessing complementary modality information. Together, these findings reveal when and how multimodal fusion would enhance biosignal classification and also provides mechanistic insights of human muscle activities. The study would be beneficial in the design of sensor arrays for neurorobotic systems.

Eric Shah, Jay Patel, Mr. Vishal Katheriya et al.

Fundus images are widely used for diagnosing various eye diseases, such as diabetic retinopathy, glaucoma, and age-related macular degeneration. However, manual analysis of fundus images is time-consuming and prone to errors. In this report, we propose a novel method for fundus detection using object detection and machine learning classification techniques. We use a YOLO_V8 to perform object detection on fundus images and locate the regions of interest (ROIs) such as optic disc, optic cup and lesions. We then use machine learning SVM classification algorithms to classify the ROIs into different DR stages based on the presence or absence of pathological signs such as exudates, microaneurysms, and haemorrhages etc. Our method achieves 84% accuracy and efficiency for fundus detection and can be applied for retinal fundus disease triage, especially in remote areas around the world.

Soumya Parekh, Jay Patel

Slating a product for release often involves pitching journalists to run stories on your press release. Good media coverage often ensures greater product reach and drives audience engagement for those products. Hence, ensuring that those releases are pitched to the right journalists with relevant interests is crucial, since they receive several pitches daily. Keeping up with journalist beats and curating a media contacts list is often a huge and time-consuming task. This study proposes a model to automate and expedite the process by recommending suitable journalists to run media coverage on the press releases provided by the user.

Sharut Gupta, Praveer Singh, Ken Chang et al.

Model brittleness is a key concern when deploying deep learning models in real-world medical settings. A model that has high performance at one institution may suffer a significant decline in performance when tested at other institutions. While pooling datasets from multiple institutions and retraining may provide a straightforward solution, it is often infeasible and may compromise patient privacy. An alternative approach is to fine-tune the model on subsequent institutions after training on the original institution. Notably, this approach degrades model performance at the original institution, a phenomenon known as catastrophic forgetting. In this paper, we develop an approach to address catastrophic forget-ting based on elastic weight consolidation combined with modulation of batch normalization statistics under two scenarios: first, for expanding the domain from one imaging system's data to another imaging system's, and second, for expanding the domain from a large multi-institutional dataset to another single institution dataset. We show that our approach outperforms several other state-of-the-art approaches and provide theoretical justification for the efficacy of batch normalization modulation. The results of this study are generally applicable to the deployment of any clinical deep learning model which requires domain expansion.

Sharut Gupta, Praveer Singh, Ken Chang et al.

Model brittleness is a primary concern when deploying deep learning models in medical settings owing to inter-institution variations, like patient demographics and intra-institution variation, such as multiple scanner types. While simply training on the combined datasets is fraught with data privacy limitations, fine-tuning the model on subsequent institutions after training it on the original institution results in a decrease in performance on the original dataset, a phenomenon called catastrophic forgetting. In this paper, we investigate trade-off between model refinement and retention of previously learned knowledge and subsequently address catastrophic forgetting for the assessment of mammographic breast density. More specifically, we propose a simple yet effective approach, adapting Elastic weight consolidation (EWC) using the global batch normalization (BN) statistics of the original dataset. The results of this study provide guidance for the deployment of clinical deep learning models where continuous learning is needed for domain expansion.

Mehak Aggarwal, Nishanth Arun, Sharut Gupta et al.

While success of Deep Learning (DL) in automated diagnosis can be transformative to the medicinal practice especially for people with little or no access to doctors, its widespread acceptability is severely limited by inherent black-box decision making and unsafe failure modes. While saliency methods attempt to tackle this problem in non-medical contexts, their apriori explanations do not transfer well to medical usecases. With this study we validate a model design element agnostic to both architecture complexity and model task, and show how introducing this element gives an inherently self-explanatory model. We compare our results with state of the art non-trainable saliency maps on RSNA Pneumonia Dataset and demonstrate a much higher localization efficacy using our adopted technique. We also compare, with a fully supervised baseline and provide a reasonable alternative to it's high data labelling overhead. We further investigate the validity of our claims through qualitative evaluation from an expert reader.

Nishanth Thumbavanam Arun, Nathan Gaw, Praveer Singh et al.

Saliency maps have become a widely used method to assess which areas of the input image are most pertinent to the prediction of a trained neural network. However, in the context of medical imaging, there is no study to our knowledge that has examined the efficacy of these techniques and quantified them using overlap with ground truth bounding boxes. In this work, we explored the credibility of the various existing saliency map methods on the RSNA Pneumonia dataset. We found that GradCAM was the most sensitive to model parameter and label randomization, and was highly agnostic to model architecture.

Katharina Hoebel, Ken Chang, Jay Patel et al.

Segmentation tasks in medical imaging are inherently ambiguous: the boundary of a target structure is oftentimes unclear due to image quality and biological factors. As such, predicted segmentations from deep learning algorithms are inherently ambiguous. Additionally, "ground truth" segmentations performed by human annotators are in fact weak labels that further increase the uncertainty of outputs of supervised models developed on these manual labels. To date, most deep learning segmentation studies utilize predicted segmentations without uncertainty quantification. In contrast, we explore the use of Monte Carlo dropout U-Nets for the segmentation with additional quantification of segmentation uncertainty. We assess the utility of three measures of uncertainty (Coefficient of Variation, Mean Pairwise Dice, and Mean Voxelwise Uncertainty) for the segmentation of a less ambiguous target structure (liver) and a more ambiguous one (liver tumors). Furthermore, we assess how the utility of these measures changes with different patch sizes and cost functions. Our results suggest that models trained using larger patches and the weighted categorical cross-entropy as cost function allow the extraction of more meaningful uncertainty measures compared to smaller patches and soft dice loss. Among the three uncertainty measures Mean Pairwise Dice shows the strongest correlation with segmentation quality. Our study serves as a proof-of-concept of how uncertainty measures can be used to assess the quality of a predicted segmentation, potentially serving to flag low quality segmentations from a given model for further human review.