Aman Sinha

Eli Bronstein, Sirish Srinivasan, Supratik Paul et al.

ML-based motion planning is a promising approach to produce agents that exhibit complex behaviors, and automatically adapt to novel environments. In the context of autonomous driving, it is common to treat all available training data equally. However, this approach produces agents that do not perform robustly in safety-critical settings, an issue that cannot be addressed by simply adding more data to the training set - we show that an agent trained using only a 10% subset of the data performs just as well as an agent trained on the entire dataset. We present a method to predict the inherent difficulty of a driving situation given data collected from a fleet of autonomous vehicles deployed on public roads. We then demonstrate that this difficulty score can be used in a zero-shot transfer to generate curricula for an imitation-learning based planning agent. Compared to training on the entire unbiased training dataset, we show that prioritizing difficult driving scenarios both reduces collisions by 15% and increases route adherence by 14% in closed-loop evaluation, all while using only 10% of the training data.

Rohit Agarwal, Aman Sinha, Ayan Vishwakarma et al.

Modeling irregularly-sampled time series (ISTS) is challenging because of missing values. Most existing methods focus on handling ISTS by converting irregularly sampled data into regularly sampled data via imputation. These models assume an underlying missing mechanism, which may lead to unwanted bias and sub-optimal performance. We present SLAN (Switch LSTM Aggregate Network), which utilizes a group of LSTMs to model ISTS without imputation, eliminating the assumption of any underlying process. It dynamically adapts its architecture on the fly based on the measured sensors using switches. SLAN exploits the irregularity information to explicitly capture each sensor's local summary and maintains a global summary state throughout the observational period. We demonstrate the efficacy of SLAN on two public datasets, namely, MIMIC-III, and Physionet 2012.

Aman Sinha, Timothee Mickus, Marianne Clausel et al.

The success of pretrained language models (PLMs) across a spate of use-cases has led to significant investment from the NLP community towards building domain-specific foundational models. On the other hand, in mission critical settings such as biomedical applications, other aspects also factor in-chief of which is a model's ability to produce reasonable estimates of its own uncertainty. In the present study, we discuss these two desiderata through the lens of how they shape the entropy of a model's output probability distribution. We find that domain specificity and uncertainty awareness can often be successfully combined, but the exact task at hand weighs in much more strongly.

Ioana Buhnila, Aman Sinha, Mathieu Constant

Recent surge in the accessibility of large language models (LLMs) to the general population can lead to untrackable use of such models for medical-related recommendations. Language generation via LLMs models has two key problems: firstly, they are prone to hallucination and therefore, for any medical purpose they require scientific and factual grounding; secondly, LLMs pose tremendous challenge to computational resources due to their gigantic model size. In this work, we introduce pRAGe, a pipeline for Retrieval Augmented Generation and evaluation of medical paraphrases generation using Small Language Models (SLM). We study the effectiveness of SLMs and the impact of external knowledge base for medical paraphrase generation in French.

Michal Štefánik, Timothee Mickus, Marek Kadlčík et al.

Recent work has shown that different large language models (LLMs) converge to similar and accurate input embedding representations for numbers. These findings conflict with the documented propensity of LLMs to produce erroneous outputs when dealing with numeric information. In this work, we aim to explain this conflict by exploring how language models manipulate numbers and quantify the lower bounds of accuracy of these mechanisms. We find that despite surfacing errors, different language models learn interchangeable representations of numbers that are systematic, highly accurate and universal across their hidden states and the types of input contexts. This allows us to create universal probes for each LLM and to trace information -- including the causes of output errors -- to specific layers. Our results lay a fundamental understanding of how pre-trained LLMs manipulate numbers and outline the potential of more accurate probing techniques in addressed refinements of LLMs' architectures.

Matthew O'Kelly, Aman Sinha, Justin Norden et al.

Modern treatments for Type 1 diabetes (T1D) use devices known as artificial pancreata (APs), which combine an insulin pump with a continuous glucose monitor (CGM) operating in a closed-loop manner to control blood glucose levels. In practice, poor performance of APs (frequent hyper- or hypoglycemic events) is common enough at a population level that many T1D patients modify the algorithms on existing AP systems with unregulated open-source software. Anecdotally, the patients in this group have shown superior outcomes compared with standard of care, yet we do not understand how safe any AP system is since adverse outcomes are rare. In this paper, we construct generative models of individual patients' physiological characteristics and eating behaviors. We then couple these models with a T1D simulator approved for pre-clinical trials by the FDA. Given the ability to simulate patient outcomes in-silico, we utilize techniques from rare-event simulation theory in order to efficiently quantify the performance of a device with respect to a particular patient. We show a 72,000$\times$ speedup in simulation speed over real-time and up to 2-10 times increase in the frequency which we are able to sample adverse conditions relative to standard Monte Carlo sampling. In practice our toolchain enables estimates of the likelihood of hypoglycemic events with approximately an order of magnitude fewer simulations.

Raúl Vázquez, Timothee Mickus, Elaine Zosa et al.

We present the Mu-SHROOM shared task which is focused on detecting hallucinations and other overgeneration mistakes in the output of instruction-tuned large language models (LLMs). Mu-SHROOM addresses general-purpose LLMs in 14 languages, and frames the hallucination detection problem as a span-labeling task. We received 2,618 submissions from 43 participating teams employing diverse methodologies. The large number of submissions underscores the interest of the community in hallucination detection. We present the results of the participating systems and conduct an empirical analysis to identify key factors contributing to strong performance in this task. We also emphasize relevant current challenges, notably the varying degree of hallucinations across languages and the high annotator disagreement when labeling hallucination spans.

Ioana Buhnila, Aman Sinha, Mathieu Constant

Large Language Models (LLMs) have proven efficient in giving definition-type answers to user input queries. While for humans giving various types of answers, such as examples and paraphrases, is an easy task, LLMs struggle to provide correct answers for other than definition-type queries. In this study, we evaluated this drop in performance using TrackList, a fine-grained linguistic and statistical analysis pipeline to investigate the impact of the pre-training data on LLMs answers to diverse linguistic queries. We also introduce RefoMed-EN, an English dataset consisting of 6170 human-annotated medical terms alongside their corresponding definitions, denominations, exemplifications, explanations, or paraphrases. We studied whether the high frequency of a concept (head) or low frequency (tail) impacts the language model's performance. We evaluated the quality of the LLM's output using syntactic and semantic similarity metrics, statistical correlations and embeddings. Results showed that the LLM's task performance for definition type questions is the highest, while for the exemplification type it is the lowest. Additionally, we showed that for definition-type questions, large language models are prone to paraphrase more on popular and frequent knowledge and less on tail and technical knowledge, especially in the expert texts.

Federica Gamba, Aman Sinha, Timothee Mickus et al.

We introduce the CAP (Confabulations from ACL Publications) dataset, a multilingual resource for studying hallucinations in large language models (LLMs) within scientific text generation. CAP focuses on the scientific domain, where hallucinations can distort factual knowledge, as they frequently do. In this domain, however, the presence of specialized terminology, statistical reasoning, and context-dependent interpretations further exacerbates these distortions, particularly given LLMs' lack of true comprehension, limited contextual understanding, and bias toward surface-level generalization. CAP operates in a cross-lingual setting covering five high-resource languages (English, French, Hindi, Italian, and Spanish) and four low-resource languages (Bengali, Gujarati, Malayalam, and Telugu). The dataset comprises 900 curated scientific questions and over 7000 LLM-generated answers from 16 publicly available models, provided as question-answer pairs along with token sequences and corresponding logits. Each instance is annotated with a binary label indicating the presence of a scientific hallucination, denoted as a factuality error, and a fluency label, capturing issues in the linguistic quality or naturalness of the text. CAP is publicly released to facilitate advanced research on hallucination detection, multilingual evaluation of LLMs, and the development of more reliable scientific NLP systems.

Aman Sinha, Bogdan-Valentin Popescu, Xavier Coubez et al.

Language models (LMs) capabilities have grown with a fast pace over the past decade leading researchers in various disciplines, such as biomedical research, to increasingly explore the utility of LMs in their day-to-day applications. Domain specific language models have already been in use for biomedical natural language processing (NLP) applications. Recently however, the interest has grown towards medical language models and their understanding capabilities. In this paper, we investigate the medical conceptual grounding of various language models against expert clinicians for identification of hallmarks of immunotherapy in breast cancer abstracts. Our results show that pre-trained language models have potential to outperform large language models in identifying very specific (low-level) concepts.

Timothee Mickus, Aman Sinha, Raúl Vázquez

The difficulty intrinsic to a given example, rooted in its inherent ambiguity, is a key yet often overlooked factor in evaluating neural NLP models. We investigate the interplay and divergence among various metrics for assessing intrinsic difficulty, including annotator dissensus, training dynamics, and model confidence. Through a comprehensive analysis using 29 models on three datasets, we reveal that while correlations exist among these metrics, their relationships are neither linear nor monotonic. By disentangling these dimensions of uncertainty, we aim to refine our understanding of data complexity and its implications for evaluating and improving NLP models.

Aman Sinha, Payam Nikdel, Supratik Paul et al.

Ensuring the safety of autonomous vehicles (AVs) requires both accurate estimation of their performance and efficient discovery of potential failure cases. This paper introduces Bayesian adaptive multifidelity sampling (BAMS), which leverages the power of adaptive Bayesian sampling to achieve efficient discovery while simultaneously estimating the rate of adverse events. BAMS prioritizes exploration of regions with potentially low performance, leading to the identification of novel and critical scenarios that traditional methods might miss. Using real-world AV data we demonstrate that BAMS discovers 10 times as many issues as Monte Carlo (MC) and importance sampling (IS) baselines, while at the same time generating rate estimates with variances 15 and 6 times narrower than MC and IS baselines respectively.

Natenaile Asmamaw Shiferaw, Simpenzwe Honore Leandre, Aman Sinha et al.

Course Outcome (CO) and Program Outcome (PO)/Program-Specific Outcome (PSO) alignment is a crucial task for ensuring curriculum coherence and assessing educational effectiveness. The construction of a Course Articulation Matrix (CAM), which quantifies the relationship between COs and POs/PSOs, typically involves assigning numerical values (0, 1, 2, 3) to represent the degree of alignment. In this study, We experiment with four models from the BERT family: BERT Base, DistilBERT, ALBERT, and RoBERTa, and use multiclass classification to assess the alignment between CO and PO/PSO pairs. We first evaluate traditional machine learning classifiers, such as Decision Tree, Random Forest, and XGBoost, and then apply transfer learning to evaluate the performance of the pretrained BERT models. To enhance model interpretability, we apply Explainable AI technique, specifically Local Interpretable Model-agnostic Explanations (LIME), to provide transparency into the decision-making process. Our system achieves accuracy, precision, recall, and F1-score values of 98.66%, 98.67%, 98.66%, and 98.66%, respectively. This work demonstrates the potential of utilizing transfer learning with BERT-based models for the automated generation of CAMs, offering high performance and interpretability in educational outcome assessment.

Deepanshu Pandey, Pradyumna Gupta, Sumit Bhattacharya et al.

Region proposal based methods like R-CNN and Faster R-CNN models have proven to be extremely successful in object detection and segmentation tasks. Recently, Transformers have also gained popularity in the domain of Computer Vision, and are being utilised to improve the performance of conventional models. In this paper, we present a relatively new transformer based approach to enhance the performance of the conventional convolutional feature extractor in the existing region proposal based methods. Our approach merges the convolutional feature maps with transformer-based token embeddings by applying a projection operation similar to self-attention in transformers. The results of our experiments show that transformer assisted feature extractor achieves a significant improvement in mIoU (mean Intersection over Union) scores compared to vanilla convolutional backbone.

Sirigireddy Dhanalaxmi, Rohit Agarwal, Aman Sinha

Social media such as Twitter is a hotspot of user-generated information. In this ongoing Covid-19 pandemic, there has been an abundance of data on social media which can be classified as informative and uninformative content. In this paper, we present our work to detect informative Covid-19 English tweets using RoBERTa model as a part of the W-NUT workshop 2020. We show the efficacy of our model on a public dataset with an F1-score of 0.89 on the validation dataset and 0.87 on the leaderboard.

Aadarsh Singh, Priyanshu Kumar, Aman Sinha

We explore the performance of Bidirectional Encoder Representations from Transformers (BERT) at definition extraction. We further propose a joint model of BERT and Text Level Graph Convolutional Network so as to incorporate dependencies into the model. Our proposed model produces better results than BERT and achieves comparable results to BERT with fine tuned language model in DeftEval (Task 6 of SemEval 2020), a shared task of classifying whether a sentence contains a definition or not (Subtask 1).

Aman Sinha, Matthew O'Kelly, Russ Tedrake et al.

Learning-based methodologies increasingly find applications in safety-critical domains like autonomous driving and medical robotics. Due to the rare nature of dangerous events, real-world testing is prohibitively expensive and unscalable. In this work, we employ a probabilistic approach to safety evaluation in simulation, where we are concerned with computing the probability of dangerous events. We develop a novel rare-event simulation method that combines exploration, exploitation, and optimization techniques to find failure modes and estimate their rate of occurrence. We provide rigorous guarantees for the performance of our method in terms of both statistical and computational efficiency. Finally, we demonstrate the efficacy of our approach on a variety of scenarios, illustrating its usefulness as a tool for rapid sensitivity analysis and model comparison that are essential to developing and testing safety-critical autonomous systems.

Pradyumna Gupta, Himanshu Gupta, Aman Sinha

Memes have become an ubiquitous social media entity and the processing and analysis of suchmultimodal data is currently an active area of research. This paper presents our work on theMemotion Analysis shared task of SemEval 2020, which involves the sentiment and humoranalysis of memes. We propose a system which uses different bimodal fusion techniques toleverage the inter-modal dependency for sentiment and humor classification tasks. Out of all ourexperiments, the best system improved the baseline with macro F1 scores of 0.357 on SentimentClassification (Task A), 0.510 on Humor Classification (Task B) and 0.312 on Scales of SemanticClasses (Task C).

Aman Sinha, Matthew O'Kelly, Hongrui Zheng et al.

Balancing performance and safety is crucial to deploying autonomous vehicles in multi-agent environments. In particular, autonomous racing is a domain that penalizes safe but conservative policies, highlighting the need for robust, adaptive strategies. Current approaches either make simplifying assumptions about other agents or lack robust mechanisms for online adaptation. This work makes algorithmic contributions to both challenges. First, to generate a realistic, diverse set of opponents, we develop a novel method for self-play based on replica-exchange Markov chain Monte Carlo. Second, we propose a distributionally robust bandit optimization procedure that adaptively adjusts risk aversion relative to uncertainty in beliefs about opponents' behaviors. We rigorously quantify the tradeoffs in performance and robustness when approximating these computations in real-time motion-planning, and we demonstrate our methods experimentally on autonomous vehicles that achieve scaled speeds comparable to Formula One racecars.

Justin Norden, Matthew O'Kelly, Aman Sinha

While autonomous vehicle (AV) technology has shown substantial progress, we still lack tools for rigorous and scalable testing. Real-world testing, the $\textit{de-facto}$ evaluation method, is dangerous to the public. Moreover, due to the rare nature of failures, billions of miles of driving are needed to statistically validate performance claims. Thus, the industry has largely turned to simulation to evaluate AV systems. However, having a simulation stack alone is not a solution. A simulation testing framework needs to prioritize which scenarios to run, learn how the chosen scenarios provide coverage of failure modes, and rank failure scenarios in order of importance. We implement a simulation testing framework that evaluates an entire modern AV system as a black box. This framework estimates the probability of accidents under a base distribution governing standard traffic behavior. In order to accelerate rare-event probability evaluation, we efficiently learn to identify and rank failure scenarios via adaptive importance-sampling methods. Using this framework, we conduct the first independent evaluation of a full-stack commercial AV system, Comma AI's OpenPilot.

Matthew O'Kelly, Aman Sinha, Hongseok Namkoong et al.

While recent developments in autonomous vehicle (AV) technology highlight substantial progress, we lack tools for rigorous and scalable testing. Real-world testing, the $\textit{de facto}$ evaluation environment, places the public in danger, and, due to the rare nature of accidents, will require billions of miles in order to statistically validate performance claims. We implement a simulation framework that can test an entire modern autonomous driving system, including, in particular, systems that employ deep-learning perception and control algorithms. Using adaptive importance-sampling methods to accelerate rare-event probability evaluation, we estimate the probability of an accident under a base distribution governing standard traffic behavior. We demonstrate our framework on a highway scenario, accelerating system evaluation by $2$-$20$ times over naive Monte Carlo sampling methods and $10$-$300 \mathsf{P}$ times (where $\mathsf{P}$ is the number of processors) over real-world testing.

Daniel Ritter, Stefanie Rinderle-Ma, Marco Montali et al.

Enterprise application integration (EAI) solutions are the centrepiece of current enterprise IT architectures (e.g., cloud and mobile computing, business networks), however, require the formalization of their building blocks, represented by integration patterns, verification and optimization. This work serves as an instructive pattern formalization catalog that leads to the formalization of all currently known integration patterns. Therefore, we explain the classification of the underlying requirements of the pattern semantics and formalize representative patterns from the different categories, by realizing them in timed db-net. In this way, the catalog will allow for the addition of future patterns by assigning them to a category and applying the described formalism.

Aman Sinha, Hongseok Namkoong, Riccardo Volpi et al.

Neural networks are vulnerable to adversarial examples and researchers have proposed many heuristic attack and defense mechanisms. We address this problem through the principled lens of distributionally robust optimization, which guarantees performance under adversarial input perturbations. By considering a Lagrangian penalty formulation of perturbing the underlying data distribution in a Wasserstein ball, we provide a training procedure that augments model parameter updates with worst-case perturbations of training data. For smooth losses, our procedure provably achieves moderate levels of robustness with little computational or statistical cost relative to empirical risk minimization. Furthermore, our statistical guarantees allow us to efficiently certify robustness for the population loss. For imperceptible perturbations, our method matches or outperforms heuristic approaches.