Search

Small Language Models (SLMs) have become increasingly important due to their efficiency and performance to perform various language tasks with minimal computational resources, making them ideal for various settings including on-device, mobile, edge devices, among many others. In this article, we present a comprehensive survey on SLMs, focusing on their architectures, training techniques, and model compression techniques. We propose a novel taxonomy for categorizing the methods used to optimize SLMs, including model compression, pruning, and quantization techniques. We summarize the benchmark datasets that are useful for benchmarking SLMs along with the evaluation metrics commonly used. Additionally, we highlight key open challenges that remain to be addressed. Our survey aims to serve as a valuable resource for researchers and practitioners interested in developing and deploying small yet efficient language models.

With the emergence of the Metaverse and focus on wearable devices in the recent years gesture based human-computer interaction has gained significance. To enable gesture recognition for VR/AR headsets and glasses several datasets focusing on egocentric i.e. first-person view have emerged in recent years. However, standard frame-based vision suffers from limitations in data bandwidth requirements as well as ability to capture fast motions. To overcome these limitation bio-inspired approaches such as event-based cameras present an attractive alternative. In this work, we present the first event-camera based egocentric gesture dataset for enabling neuromorphic, low-power solutions for XR-centric gesture recognition. The dataset has been made available publicly at the following URL: https://gitlab.com/NVM_IITD_Research/xrage.

High-redshift protoclusters consisting of dusty starbursts are thought to play an important role in galaxy evolution. Their dusty nature makes them bright in the FIR/submm but difficult to find in optical/NIR surveys. Radio observations are an excellent way to study these dusty starbursts, as dust is transparent in the radio and there is a tight correlation between the FIR and radio emission of a galaxy. Here, we present MeerKAT 1.28 GHz radio imaging of 3 Herschel candidate protoclusters, with a synthesised beam size of ~$7.5'' \times 6.6''$ and a central thermal noise down to $4.35~\mu$Jy/beam. Our source counts are consistent with other radio counts with no evidence of overdensities. Around 95% of the Herschel sources have 1.28 GHz IDs. Using the Herschel 250 micron primary beam size as the searching radius, we find 54.2% Herschel sources have multiple 1.28 GHz IDs. Our average FIR-radio correlation coefficient $q_{250\mu m}$ is $2.33\pm 0.26$. Adding $q_{250\mu m}$ as a new constraint, the probability of finding chance-aligned sources is reduced by a factor of ~6, but with the risk of discarding true identifications of radio-loud/quiet sources. With accurate MeerKAT positions, we cross-match our Herschel sources to optical/NIR data followed by photometric redshift estimations. By removing $z<1$ sources, the density contrasts of two of the candidate protoclusters increase, suggestive of them being real protoclusters at $z>1$. There is also potentially a $0.9

Multi-modal Large Language Models (MLLMs) exhibit impressive problem-solving abilities in various domains, but their visual comprehension and abstract reasoning skills remain under-evaluated. To this end, we present PolyMATH, a challenging benchmark aimed at evaluating the general cognitive reasoning abilities of MLLMs. PolyMATH comprises 5,000 manually collected high-quality images of cognitive textual and visual challenges across 10 distinct categories, including pattern recognition, spatial reasoning, and relative reasoning. We conducted a comprehensive, and quantitative evaluation of 15 MLLMs using four diverse prompting strategies, including Chain-of-Thought and Step-Back. The best scores achieved on PolyMATH are ~41%, ~36%, and ~27%, obtained by Claude-3.5 Sonnet, GPT-4o and Gemini-1.5 Pro respectively - highlighting the logical and visual complexity of these questions. A further fine-grained error analysis reveals that these models struggle to understand spatial relations and perform drawn-out, high-level reasoning. This is further strengthened by our ablation study estimating MLLM performance when given textual descriptions in place of diagrams. As evidenced by ~4% improvement over textual descriptions as opposed to actual images, we discover that models do not truly comprehend visual diagrams and the spatial information therein, and are thus prone to logical errors. Finally, we evaluate the OpenAI o1 models and find that their performance only matches the human baseline, highlighting the difficulty of the benchmark. The results on PolyMATH highlight the room for improvement in multi-modal reasoning and provide unique insights to guide the development of future MLLMs., Comment: 49 pages, (10 pages paper, 9 pages references, 30 pages appendix)

Accurate assessment of disease severity from endoscopy videos in ulcerative colitis (UC) is crucial for evaluating drug efficacy in clinical trials. Severity is often measured by the Mayo Endoscopic Subscore (MES) and Ulcerative Colitis Endoscopic Index of Severity (UCEIS) score. However, expert MES/UCEIS annotation is time-consuming and susceptible to inter-rater variability, factors addressable by automation. Automation attempts with frame-level labels face challenges in fully-supervised solutions due to the prevalence of video-level labels in clinical trials. CNN-based weakly-supervised models (WSL) with end-to-end (e2e) training lack generalization to new disease scores and ignore spatio-temporal information crucial for accurate scoring. To address these limitations, we propose "Arges", a deep learning framework that utilizes a transformer with positional encoding to incorporate spatio-temporal information from frame features to estimate disease severity scores in endoscopy video. Extracted features are derived from a foundation model (ArgesFM), pre-trained on a large diverse dataset from multiple clinical trials (61M frames, 3927 videos). We evaluate four UC disease severity scores, including MES and three UCEIS component scores. Test set evaluation indicates significant improvements, with F1 scores increasing by 4.1% for MES and 18.8%, 6.6%, 3.8% for the three UCEIS component scores compared to state-of-the-art methods. Prospective validation on previously unseen clinical trial data further demonstrates the model's successful generalization., Comment: 12 pages, 2 figures, 5 tables, accepted at MLMI, MICCAI

In this paper, we employ the Density Dependent Relativistic Hadron (DDRH) field theoretical Model in a Bayesian analysis to investigate the equation of state (EOS) of dense matter featuring antikaon condensation for $K^-$ and $\bar{K}^0$ inside neutron stars. The vector coupling parameters within the kaonic sector are determined through the iso-spin counting rule and quark model. Our study integrates various constraints, including $\chi$EFT calculations, nuclear saturation properties, and astrophysical observations from pulsars PSR J0030+0451 and PSR J0740+66 and from the GW170817 event. We present posterior distributions of model parameters derived from these constraints, enabling us to explore the distributions of nuclear matter properties and neutron star (NS) characteristics such as radii, tidal deformabilities, central energy densities, and speed of sound. The antikaon potential at the 68(90)\% confidence intervals is determined to be $-129.36^{+12.53(+32.617)}_{-3.837(-5.696)}$ MeV. This aligns with several studies providing estimates within the range of $-120$ to $-150$ MeV. We find that the maximum neutron star mass is constrained to around 2M$_\odot$ due to the significant softening of the EOS caused by antikaon condensation. This softening results in a considerable decrease in the speed of sound. Although antikaon condensation for $K^-$ is not feasible inside the canonical neutron stars, it becomes feasible for higher NS masses. The condensation of both $K^-$ and $\bar{K}^0$ is probably present in the interior of neutron star with mass greater than 2M$_\odot$. We also discuss the interconnections among input variables, isoscalar and isovector aspects of the EOS, and specific NS properties in the context of antikaon condensation., Comment: 19 pages, 10 figures, Accepted for publication in Physical Review C (in press)

Metallic glasses have mechanical properties, which exhibit avalanches in the disguise of stress drops. We study these phenomena in a classical metallic glass system CuZrAl by athermal quasistatic shear and varying the element concentrations and for pure CuZr 50/50 case the cooling rate. The resulting mechanical properties are close to the behaviour found experimentally. At small strains, the pristine systems are akin to other glassy systems with a so-called gap distribution with a small positive exponent. Critical avalanching behaviour is found only approaching the yield point. The post-yield stress drops are universal, and the gap distribution becomes flat.

Time Series Foundation Models (TSFMs) have recently garnered attention for their ability to model complex, large-scale time series data across domains such as retail, finance, and transportation. However, their application to sensitive, domain-specific fields like healthcare remains challenging, primarily due to the difficulty of fine-tuning these models for specialized, out-of-domain tasks with scarce publicly available datasets. In this work, we explore the use of Parameter-Efficient Fine-Tuning (PEFT) techniques to address these limitations, focusing on healthcare applications, particularly ICU vitals forecasting for sepsis patients. We introduce and evaluate two selective (BitFit and LayerNorm Tuning) and two additive (VeRA and FourierFT) PEFT techniques on multiple configurations of the Chronos TSFM for forecasting vital signs of sepsis patients. Our comparative analysis demonstrates that some of these PEFT methods outperform LoRA in terms of parameter efficiency and domain adaptation, establishing state-of-the-art (SOTA) results in ICU vital forecasting tasks. Interestingly, FourierFT applied to the Chronos (Tiny) variant surpasses the SOTA model while fine-tuning only 2,400 parameters compared to the 700K parameters of the benchmark., Comment: 7 pages. Under review

Open-source, multilingual medical large language models (LLMs) have the potential to serve linguistically diverse populations across different regions. Adapting generic LLMs for healthcare often requires continual pretraining, but this approach is computationally expensive and sometimes impractical. Instruction fine-tuning on a specific task may not always guarantee optimal performance due to the lack of broader domain knowledge that the model needs to understand and reason effectively in diverse scenarios. To address these challenges, we introduce two multilingual instruction fine-tuning datasets, MMed-IFT and MMed-IFT-MC, containing over 200k high-quality medical samples in six languages. We propose a two-stage training paradigm: the first stage injects general medical knowledge using MMed-IFT, while the second stage fine-tunes task-specific multiple-choice questions with MMed-IFT-MC. Our method achieves competitive results on both English and multilingual benchmarks, striking a balance between computational efficiency and performance. We plan to make our dataset and model weights public at \url{https://github.com/SpassMed/Med-Llama3} in the future., Comment: Technical Report v1, work in progress

LLMs demonstrate an uncanny ability to process unstructured data, and as such, have the potential to go beyond search and run complex, semantic analyses at scale. We describe the design of an unstructured analytics system, Aryn, and the tenets and use cases that motivate its design. With Aryn, users can specify queries in natural language and the system automatically determines a semantic plan and executes it to compute an answer from a large collection of unstructured documents using LLMs. At the core of Aryn is Sycamore, a declarative document processing engine, built using Ray, that provides a reliable distributed abstraction called DocSets. Sycamore allows users to analyze, enrich, and transform complex documents at scale. Aryn also comprises Luna, a query planner that translates natural language queries to Sycamore scripts, and the Aryn Partitioner, which takes raw PDFs and document images, and converts them to DocSets for downstream processing. Using Aryn, we demonstrate a real world use case for analyzing accident reports from the National Transportation Safety Board (NTSB), and discuss some of the major challenges we encountered in deploying Aryn in the wild., Comment: 6 pages, 3 figures, fixed typos

The feedback between mechanical and chemical signals plays a key role in controlling many biological processes and collective cell behavior. Here we focus on the emergence of spatiotemporal density waves in a one-dimensional "cell train." Combining a minimal theoretical model with observations in an in vitro experimental system of MDCK epithelial cells confined to a linear pattern, we examine the spontaneous oscillations driven by the feedback between myosin activation and mechanical deformations and their effect on the response of the tissue to externally applied deformations. We show that the nature and frequency of spontaneous oscillations is controlled by the size of the cell train, with a transition from size-dependent standing waves to intrinsic spontaneous waves at the natural frequency of the tissue. The response to external boundary perturbations exhibit a resonance at this natural frequency, providing a possible venue for inferring the mechanochemical couplings that control the tissue behavior from rheological experiments., Comment: 15 pages, 10 figures

In the domain of black-box model extraction, conventional methods reliant on soft labels or surrogate datasets struggle with scaling to high-dimensional input spaces and managing the complexity of an extensive array of interrelated classes. In this work, we present a novel approach that utilizes SHAP (SHapley Additive exPlanations) to enhance synthetic data generation. SHAP quantifies the individual contributions of each input feature towards the victim model's output, facilitating the optimization of an energy-based GAN towards a desirable output. This method significantly boosts performance, achieving a 16.45% increase in the accuracy of image classification models and extending to video classification models with an average improvement of 26.11% and a maximum of 33.36% on challenging datasets such as UCF11, UCF101, Kinetics 400, Kinetics 600, and Something-Something V2. We further demonstrate the effectiveness and practical utility of our method under various scenarios, including the availability of top-k prediction probabilities, top-k prediction labels, and top-1 labels.

Solving grid puzzles involves a significant amount of logical reasoning. Hence, it is a good domain to evaluate the reasoning capability of a model which can then guide us to improve the reasoning ability of models. However, most existing works evaluate only the final predicted answer of a puzzle, without delving into an in-depth analysis of the LLMs' reasoning chains (such as where they falter) or providing any finer metrics to evaluate them. Since LLMs may rely on simple heuristics or artifacts to predict the final answer, it is crucial to evaluate the generated reasoning chain beyond overall correctness measures, for accurately evaluating the reasoning abilities of LLMs. To this end, we first develop GridPuzzle, an evaluation dataset comprising 274 grid-based puzzles with different complexities. Second, we propose a new error taxonomy derived from manual analysis of reasoning chains from LLMs including GPT-4, Claude-3, Gemini, Mistral, and Llama-2. Then, we develop an LLM-based framework for large-scale subjective evaluation (i.e., identifying errors) and an objective metric, PuzzleEval, to evaluate the correctness of reasoning chains. Evaluating reasoning chains from LLMs leads to several interesting findings. We further show that existing prompting methods used for enhancing models' reasoning abilities do not improve performance on GridPuzzle. This highlights the importance of understanding fine-grained errors and presents a challenge for future research to enhance LLMs' puzzle-solving abilities by developing methods that address these errors. Data and source code are available at https://github.com/Mihir3009/GridPuzzle., Comment: Accepted at EMNLP 2024 Main

The recent strides in artificial intelligence (AI) and machine learning (ML) have propelled the rise of TinyML, a paradigm enabling AI computations at the edge without dependence on cloud connections. While TinyML offers real-time data analysis and swift responses critical for diverse applications, its devices' intrinsic resource limitations expose them to security risks. This research delves into the adversarial vulnerabilities of AI models on resource-constrained embedded hardware, with a focus on Model Extraction and Evasion Attacks. Our findings reveal that adversarial attacks from powerful host machines could be transferred to smaller, less secure devices like ESP32 and Raspberry Pi. This illustrates that adversarial attacks could be extended to tiny devices, underscoring vulnerabilities, and emphasizing the necessity for reinforced security measures in TinyML deployments. This exploration enhances the comprehension of security challenges in TinyML and offers insights for safeguarding sensitive data and ensuring device dependability in AI-powered edge computing settings., Comment: Accepted and presented at tinyML Foundation EMEA Innovation Forum 2024

As language models have scaled both their number of parameters and pretraining dataset sizes, the computational cost for pretraining has become intractable except for the most well-resourced teams. This increasing cost makes it ever more important to be able to reuse a model after it has completed pretraining; allowing for a model's abilities to further improve without needing to train from scratch. In this work, we detail a set of guidelines that cover how to design efficacious data distributions and learning rate schedules for continued pretraining of language models. When applying these findings within a continued pretraining run on top of a well-trained 15B parameter model, we show an improvement of 9\% in average model accuracy compared to the baseline of continued training on the pretraining set. The resulting recipe provides a practical starting point with which to begin developing language models through reuse rather than retraining., Comment: Preprint. Under review

The impressive capabilities of recent language models can be largely attributed to the multi-trillion token pretraining datasets that they are trained on. However, model developers fail to disclose their construction methodology which has lead to a lack of open information on how to develop effective pretraining sets. To address this issue, we perform the first systematic study across the entire pipeline of pretraining set construction. First, we run ablations on existing techniques for pretraining set development to identify which methods translate to the largest gains in model accuracy on downstream evaluations. Then, we categorize the most widely used data source, web crawl snapshots, across the attributes of toxicity, quality, type of speech, and domain. Finally, we show how such attribute information can be used to further refine and improve the quality of a pretraining set. These findings constitute an actionable set of steps that practitioners can use to develop high quality pretraining sets., Comment: Accepted as an oral presentation at EMNLP 2024

Extractive summarization plays a pivotal role in natural language processing due to its wide-range applications in summarizing diverse content efficiently, while also being faithful to the original content. Despite significant advancement achieved in extractive summarization by Large Language Models (LLMs), these summaries frequently exhibit incoherence. An important aspect of the coherent summary is its readability for intended users. Although there have been many datasets and benchmarks proposed for creating coherent extractive summaries, none of them currently incorporate user intent to improve coherence in extractive summarization. Motivated by this, we propose a systematically created human-annotated dataset consisting of coherent summaries for five publicly available datasets and natural language user feedback, offering valuable insights into how to improve coherence in extractive summaries. We utilize this dataset for aligning LLMs through supervised fine-tuning with natural language human feedback to enhance the coherence of their generated summaries. Preliminary experiments with Falcon-40B and Llama-2-13B show significant performance improvements (~10% Rouge-L) in terms of producing coherent summaries. We further utilize human feedback to benchmark results over instruction-tuned models such as FLAN-T5 which resulted in several interesting findings. Data and source code are available at https://github.com/Mihir3009/Extract-AI., Comment: 10 pages

Clay, the major ingredient of natural soils, is often used as a rheological modifier while formulating paints and coatings. When subjected to desiccation, colloidal clay suspensions and clayey soils crack due to the accumulation of drying-induced stresses. Even when desiccation is suppressed, aqueous clay suspensions exhibit physical aging, with their elastic and viscous moduli increasing over time as their microscopic structures evolve due to time-dependent inter-particle screened electrostatic interactions. The rate at which aging progresses is estimated from the rate of evolution of the mechanical moduli and can be controlled by changing clay concentration or by incorporating additives. Since physical aging and evaporation should both contribute to the consolidation of drying clay suspensions, we attempt to manipulate the desiccation process \textit{via} alterations of clay and additive concentrations. For a desiccating sample with an accelerated rate of aging, we observe faster consolidation into a semi-solid state and earlier onset of cracks. We estimate the crack onset time, $t_c$, in direct visualization experiments and the elasticity of the drying sample layer, $E$, using microindentation in an atomic force microscope. We demonstrate that $t_c \propto \sqrt{\frac{G_c}{E}}$, where $G_c$, the fracture energy, is estimated by fitting our experimental data to a linear poroelastic model that incorporates the Griffith's criterion for crack formation. Our work demonstrates that early crack onset is associated with lower sample ductility. The correlation between crack onset in a sample and its mechanical properties as uncovered here is potentially useful in preparing crack-resistant coatings and diverse clay structures., Comment: 9 pages, 6 figures, Supplementary Information file and supplementary videos are provided

As Large Language Models (LLMs) continue to exhibit remarkable performance in natural language understanding tasks, there is a crucial need to measure their ability for human-like multi-step logical reasoning. Existing logical reasoning evaluation benchmarks often focus primarily on simplistic single-step or multi-step reasoning with a limited set of inference rules. Furthermore, the lack of datasets for evaluating non-monotonic reasoning represents a crucial gap since it aligns more closely with human-like reasoning. To address these limitations, we propose Multi-LogiEval, a comprehensive evaluation dataset encompassing multi-step logical reasoning with various inference rules and depths. Multi-LogiEval covers three logic types--propositional, first-order, and non-monotonic--consisting of more than 30 inference rules and more than 60 of their combinations with various depths. Leveraging this dataset, we conduct evaluations on a range of LLMs including GPT-4, ChatGPT, Gemini-Pro, Yi, Orca, and Mistral, employing a zero-shot chain-of-thought. Experimental results show that there is a significant drop in the performance of LLMs as the reasoning steps/depth increases (average accuracy of ~68% at depth-1 to ~43% at depth-5). We further conduct a thorough investigation of reasoning chains generated by LLMs which reveals several important findings. We believe that Multi-LogiEval facilitates future research for evaluating and enhancing the logical reasoning ability of LLMs. Data is available at https://github.com/Mihir3009/Multi-LogiEval., Comment: Accepted at EMNLP 2024 Main

We release the Nemotron-4 340B model family, including Nemotron-4-340B-Base, Nemotron-4-340B-Instruct, and Nemotron-4-340B-Reward. Our models are open access under the NVIDIA Open Model License Agreement, a permissive model license that allows distribution, modification, and use of the models and its outputs. These models perform competitively to open access models on a wide range of evaluation benchmarks, and were sized to fit on a single DGX H100 with 8 GPUs when deployed in FP8 precision. We believe that the community can benefit from these models in various research studies and commercial applications, especially for generating synthetic data to train smaller language models. Notably, over 98% of data used in our model alignment process is synthetically generated, showcasing the effectiveness of these models in generating synthetic data. To further support open research and facilitate model development, we are also open-sourcing the synthetic data generation pipeline used in our model alignment process.

Because of their extreme densities and consequently, gravitational potential, compact objects such as neutron stars can prove to be excellent captors of dark matter particles. Considering purely gravitational interactions between dark and hadronic matter, we construct dark matter admixed stars composed of two-fluid matter subject to current astrophysical constraints of maximum mass and tidal deformability. We choose a wide range of parameters to construct the dark matter equation of state, and the DDME2 parameterization for the hadronic equation of state. We then examine the effect of dark matter on the stellar structure, tidal deformability and non-radial modes considering the relativistic Cowling approximation. We find the effect on $p$-modes is substantial, with frequencies decreasing up to the typical $f-$mode frequency range for most stars with a dark matter halo. The effects on the $f-$mode frequency are less extreme. Finally, we find the most probable and $1\sigma$ values of the dark matter parameters used in this study., Comment: 23 pages and 5 figures

The current standard model of cosmology successfully describes a variety of measurements, but the nature of its main ingredients, dark matter and dark energy, remains unknown. Euclid is a medium-class mission in the Cosmic Vision 2015-2025 programme of the European Space Agency (ESA) that will provide high-resolution optical imaging, as well as near-infrared imaging and spectroscopy, over about 14,000 deg^2 of extragalactic sky. In addition to accurate weak lensing and clustering measurements that probe structure formation over half of the age of the Universe, its primary probes for cosmology, these exquisite data will enable a wide range of science. This paper provides a high-level overview of the mission, summarising the survey characteristics, the various data-processing steps, and data products. We also highlight the main science objectives and expected performance., Comment: Accepted for publication in the A&A special issue`Euclid on Sky'

Advancements in wireless ultrasound technology allow for point of care cartilage imaging, yet validation against traditional ultrasound units remains to be established for knee cartilage outcomes. Therefore, the purpose of our study was to establish the agreement of articular cartilage thickness and echo-intensity measures between traditional and wireless ultrasound units utilizing automatic-gain and normalization techniques. We used traditional and wireless ultrasound to assess the femoral cartilage via transverse suprapatellar scans with the knee in maximum flexion in 71 female NCAA Division I athletes (age: 20.0$\pm$1.3 years, height: 171.7$\pm$8.7 cm, mass: 69.4$\pm$11.0 kg). Wireless ultrasound images (auto-gain and standard gain) were compared to traditional ultrasound images (standard gain) before and after normalization. Ultrasound image pixel values were algebraically scaled to normalize imaging parameter differences between units. Mean thickness and echo-intensity of the global and sub-regions of interest were measured for unnormalized and normalized images. Intraclass correlation coefficients ($ICC_{2,k}$) for absolute agreement, standard error of the measurement, and minimum detectable difference were calculated between the traditional and wireless ultrasound units across both gain parameters and normalization. Cartilage thickness demonstrated good to excellent agreement for all regions ($ICC_{2,k} = 0.83 {\text -} 0.95$) regardless of gain and normalization. However, mean echo-intensity demonstrated poor to moderate agreement in all regions regardless of gain and normalization ($ICC_{2,k} = 0.23 {\text -} 0.68 $). While there was a high level of agreement between units when assessing cartilage thickness, further research in ultrasound beam forming may lead to improvements in agreement of cartilage echo-intensity between ultrasound units., Comment: 19 Pages, 1 figure, 3 tables

Low-Rank Adaptation (LoRA) is a widely used technique for fine-tuning large pre-trained or foundational models across different modalities and tasks. However, its application to time series data, particularly within foundational models, remains underexplored. This paper examines the impact of LoRA on contemporary time series foundational models: Lag-Llama, MOIRAI, and Chronos. We demonstrate LoRA's fine-tuning potential for forecasting the vital signs of sepsis patients in intensive care units (ICUs), emphasizing the models' adaptability to previously unseen, out-of-domain modalities. Integrating LoRA aims to enhance forecasting performance while reducing inefficiencies associated with fine-tuning large models on limited domain-specific data. Our experiments show that LoRA fine-tuning of time series foundational models significantly improves forecasting, achieving results comparable to state-of-the-art models trained from scratch on similar modalities. We conduct comprehensive ablation studies to demonstrate the trade-offs between the number of tunable parameters and forecasting performance and assess the impact of varying LoRA matrix ranks on model performance., Comment: 5 pages, 3 figures. This work has been submitted to the ACM for possible publication

Sepsis is a leading cause of mortality in intensive care units (ICUs), representing a substantial medical challenge. The complexity of analyzing diverse vital signs to predict sepsis further aggravates this issue. While deep learning techniques have been advanced for early sepsis prediction, their 'black-box' nature obscures the internal logic, impairing interpretability in critical settings like ICUs. This paper introduces a framework that combines a deep learning model with an attention mechanism that highlights the critical time steps in the forecasting process, thus improving model interpretability and supporting clinical decision-making. We show that the attention mechanism could be adapted to various black box time series forecasting models such as N-HiTS and N-BEATS. Our method preserves the accuracy of conventional deep learning models while enhancing interpretability through attention-weight-generated heatmaps. We evaluated our model on the eICU-CRD dataset, focusing on forecasting vital signs for sepsis patients. We assessed its performance using mean squared error (MSE) and dynamic time warping (DTW) metrics. We explored the attention maps of N-HiTS and N-BEATS, examining the differences in their performance and identifying crucial factors influencing vital sign forecasting., Comment: 8 pages, 4 figures

Emerging vulnerabilities in machine learning (ML) models due to adversarial attacks raise concerns about their reliability. Specifically, evasion attacks manipulate models by introducing precise perturbations to input data, causing erroneous predictions. To address this, we propose a methodology combining SHapley Additive exPlanations (SHAP) for feature importance analysis with an innovative Optimal Epsilon technique for conducting evasion attacks. Our approach begins with SHAP-based analysis to understand model vulnerabilities, crucial for devising targeted evasion strategies. The Optimal Epsilon technique, employing a Binary Search algorithm, efficiently determines the minimum epsilon needed for successful evasion. Evaluation across diverse machine learning architectures demonstrates the technique's precision in generating adversarial samples, underscoring its efficacy in manipulating model outcomes. This study emphasizes the critical importance of continuous assessment and monitoring to identify and mitigate potential security risks in machine learning systems.

Recently developed large language models (LLMs) have been shown to perform remarkably well on a wide range of language understanding tasks. But, can they really "reason" over the natural language? This question has been receiving significant research attention and many reasoning skills such as commonsense, numerical, and qualitative have been studied. However, the crucial skill pertaining to 'logical reasoning' has remained underexplored. Existing work investigating this reasoning ability of LLMs has focused only on a couple of inference rules (such as modus ponens and modus tollens) of propositional and first-order logic. Addressing the above limitation, we comprehensively evaluate the logical reasoning ability of LLMs on 25 different reasoning patterns spanning over propositional, first-order, and non-monotonic logics. To enable systematic evaluation, we introduce LogicBench, a natural language question-answering dataset focusing on the use of a single inference rule. We conduct detailed analysis with a range of LLMs such as GPT-4, ChatGPT, Gemini, Llama-2, and Mistral using chain-of-thought prompting. Experimental results show that existing LLMs do not fare well on LogicBench; especially, they struggle with instances involving complex reasoning and negations. Furthermore, they sometimes overlook contextual information necessary for reasoning to arrive at the correct conclusion. We believe that our work and findings facilitate future research for evaluating and enhancing the logical reasoning ability of LLMs. Data and code are available at https://github.com/Mihir3009/LogicBench., Comment: Accepted at ACL(Main) 2024 | First version available @ https://openreview.net/forum?id=7NR2ZVzZxx

Fr\'echet Video Distance (FVD), a prominent metric for evaluating video generation models, is known to conflict with human perception occasionally. In this paper, we aim to explore the extent of FVD's bias toward per-frame quality over temporal realism and identify its sources. We first quantify the FVD's sensitivity to the temporal axis by decoupling the frame and motion quality and find that the FVD increases only slightly with large temporal corruption. We then analyze the generated videos and show that via careful sampling from a large set of generated videos that do not contain motions, one can drastically decrease FVD without improving the temporal quality. Both studies suggest FVD's bias towards the quality of individual frames. We further observe that the bias can be attributed to the features extracted from a supervised video classifier trained on the content-biased dataset. We show that FVD with features extracted from the recent large-scale self-supervised video models is less biased toward image quality. Finally, we revisit a few real-world examples to validate our hypothesis., Comment: CVPR 2024. Project webpage: https://content-debiased-fvd.github.io/

Learning a continuous and reliable representation of physical fields from sparse sampling is challenging and it affects diverse scientific disciplines. In a recent work, we present a novel model called MMGN (Multiplicative and Modulated Gabor Network) with implicit neural networks. In this work, we design additional studies leveraging explainability methods to complement the previous experiments and further enhance the understanding of latent representations generated by the model. The adopted methods are general enough to be leveraged for any latent space inspection. Preliminary results demonstrate the contextual information incorporated in the latent representations and their impact on the model performance. As a work in progress, we will continue to verify our findings and develop novel explainability approaches.

Causal video question answering (QA) has garnered increasing interest, yet existing datasets often lack depth in causal reasoning. To address this gap, we capitalize on the unique properties of cartoons and construct CausalChaos!, a novel, challenging causal Why-QA dataset built upon the iconic "Tom and Jerry" cartoon series. Cartoons use the principles of animation that allow animators to create expressive, unambiguous causal relationships between events to form a coherent storyline. Utilizing these properties, along with thought-provoking questions and multi-level answers (answer and detailed causal explanation), our questions involve causal chains that interconnect multiple dynamic interactions between characters and visual scenes. These factors demand models to solve more challenging, yet well-defined causal relationships. We also introduce hard incorrect answer mining, including a causally confusing version that is even more challenging. While models perform well, there is much room for improvement, especially, on open-ended answers. We identify more advanced/explicit causal relationship modeling & joint modeling of vision and language as the immediate areas for future efforts to focus upon. Along with the other complementary datasets, our new challenging dataset will pave the way for these developments in the field., Comment: Project Page: https://github.com/LUNAProject22/CausalChaos

Action quality assessment (AQA) applies computer vision to quantitatively assess the performance or execution of a human action. Current AQA approaches are end-to-end neural models, which lack transparency and tend to be biased because they are trained on subjective human judgements as ground-truth. To address these issues, we introduce a neuro-symbolic paradigm for AQA, which uses neural networks to abstract interpretable symbols from video data and makes quality assessments by applying rules to those symbols. We take diving as the case study. We found that domain experts prefer our system and find it more informative than purely neural approaches to AQA in diving. Our system also achieves state-of-the-art action recognition and temporal segmentation, and automatically generates a detailed report that breaks the dive down into its elements and provides objective scoring with visual evidence. As verified by a group of domain experts, this report may be used to assist judges in scoring, help train judges, and provide feedback to divers. Annotated training data and code: https://github.com/laurenok24/NSAQA., Comment: CVPR 2024 CVSports (Oral Presentation; 3/3 Strong Accepts) + Selected for CVPR 2024 Demos

In this work, we address two limitations of existing conditional diffusion models: their slow inference speed due to the iterative denoising process and their reliance on paired data for model fine-tuning. To tackle these issues, we introduce a general method for adapting a single-step diffusion model to new tasks and domains through adversarial learning objectives. Specifically, we consolidate various modules of the vanilla latent diffusion model into a single end-to-end generator network with small trainable weights, enhancing its ability to preserve the input image structure while reducing overfitting. We demonstrate that, for unpaired settings, our model CycleGAN-Turbo outperforms existing GAN-based and diffusion-based methods for various scene translation tasks, such as day-to-night conversion and adding/removing weather effects like fog, snow, and rain. We extend our method to paired settings, where our model pix2pix-Turbo is on par with recent works like Control-Net for Sketch2Photo and Edge2Image, but with a single-step inference. This work suggests that single-step diffusion models can serve as strong backbones for a range of GAN learning objectives. Our code and models are available at https://github.com/GaParmar/img2img-turbo., Comment: Github: https://github.com/GaParmar/img2img-turbo