Search

Malnutrition among newborns is a top public health concern in developing countries. Identification and subsequent growth monitoring are key to successful interventions. However, this is challenging in rural communities where health systems tend to be inaccessible and under-equipped, with poor adherence to protocol. Our goal is to equip health workers and public health systems with a solution for contactless newborn anthropometry in the community. We propose NurtureNet, a multi-task model that fuses visual information (a video taken with a low-cost smartphone) with tabular inputs to regress multiple anthropometry estimates including weight, length, head circumference, and chest circumference. We show that visual proxy tasks of segmentation and keypoint prediction further improve performance. We establish the efficacy of the model through several experiments and achieve a relative error of 3.9% and mean absolute error of 114.3 g for weight estimation. Model compression to 15 MB also allows offline deployment to low-cost smartphones., Comment: Accepted at CVPM Workshop at CVPR 2024

Python, as a multi-paradigm language known for its ease of integration with other languages, has gained significant attention among verification engineers recently. A Python-based verification environment capitalizes on open-source frameworks such as PyUVM providing Python-based UVM 1.2 implementation and PyVSC facilitating constrained randomization and functional coverage. These libraries play a pivotal role in expediting test development and hold promise for reducing setup costs. The goal of this paper is to evaluate the effectiveness of PyUVM verification testbenches across various design IPs, aiming for a comprehensive comparison of their features and performance metrics with the established SystemVerilog-UVM methodology., Comment: Published in DVCon U.S. 2024

Being the most widely used language across the world due to its simplicity and with 35 keywords (v3.7), Python attracts both hardware and software engineers. Python-based verification environment leverages open-source libraries such as cocotb and cocotb-coverage that enables interfacing the tesbenches with any available simulator and facilitating constrained randomization, coverage respectively. These libraries significantly ease the development of testbenches and have the potential to reduce the setup cost. The goal of this paper is to assess the effectiveness of a Python-Cocotb verification setup with design IPs and compare its features and performance metrics with the current de-facto hardware verification language i.e., SystemVerilog., Comment: Published in DVCon Europe 2023

Aggregating statistics over geographical regions is important for many applications, such as analyzing income, election results, and disease spread. However, the sensitive nature of this data necessitates strong privacy protections to safeguard individuals. In this work, we present a unified locational differential privacy (DP) framework to enable private aggregation of various data types, including one-hot encoded, boolean, float, and integer arrays, over geographical regions. Our framework employs local DP mechanisms such as randomized response, the exponential mechanism, and the Gaussian mechanism. We evaluate our approach on four datasets representing significant location data aggregation scenarios. Results demonstrate the utility of our framework in providing formal DP guarantees while enabling geographical data analysis., Comment: 10 pages, 7 figures

Underwater communication networks are increasingly popularized by various important maritime applications. However, this also leads to an increased threat landscape. This letter presents the first study that considers jamming attacks by random jammers present in the surroundings of legitimate transceivers in underwater acoustic communication systems. We investigate the impact of jamming attacks on various performance parameters of the legitimate underwater acoustic communication link. In particular, we investigate the legitimate link using stochastic geometry for important performance parameters, namely coverage probability, average rate, and energy efficiency of the link between two legitimate nodes, i.e., underwater and surface nodes. We then derive and present tractable expressions for these performance parameters. Finally, we performed a Monte Carlo simulation to validate our analysis. We plot the performance metrics against the transmit power, and jamming power for different intensities of the jammers in shallow, mid, and deep water scenarios. Results reveal that on average, jamming in deep water has a relatively high impact on the performance of legitimate link than in shallow water.

As AI systems become increasingly prevalent and impactful, the need for effective AI governance and accountability measures is paramount. This paper examines the AI governance landscape, focusing on Anthropic's Claude, a foundational AI model. We analyze Claude through the lens of the NIST AI Risk Management Framework and the EU AI Act, identifying potential threats and proposing mitigation strategies. The paper highlights the importance of transparency, rigorous benchmarking, and comprehensive data handling processes in ensuring the responsible development and deployment of AI systems. We conclude by discussing the social impact of AI governance and the ethical considerations surrounding AI accountability.

Machine learning (ML) techniques have been applied to high-level synthesis (HLS) flows for quality-of-result (QoR) prediction and design space exploration (DSE). Nevertheless, the scarcity of accessible high-quality HLS datasets and the complexity of building such datasets present challenges. Existing datasets have limitations in terms of benchmark coverage, design space enumeration, vendor extensibility, or lack of reproducible and extensible software for dataset construction. Many works also lack user-friendly ways to add more designs, limiting wider adoption of such datasets. In response to these challenges, we introduce HLSFactory, a comprehensive framework designed to facilitate the curation and generation of high-quality HLS design datasets. HLSFactory has three main stages: 1) a design space expansion stage to elaborate single HLS designs into large design spaces using various optimization directives across multiple vendor tools, 2) a design synthesis stage to execute HLS and FPGA tool flows concurrently across designs, and 3) a data aggregation stage for extracting standardized data into packaged datasets for ML usage. This tripartite architecture ensures broad design space coverage via design space expansion and supports multiple vendor tools. Users can contribute to each stage with their own HLS designs and synthesis results and extend the framework itself with custom frontends and tool flows. We also include an initial set of built-in designs from common HLS benchmarks curated open-source HLS designs. We showcase the versatility and multi-functionality of our framework through six case studies: I) Design space sampling; II) Fine-grained parallelism backend speedup; III) Targeting Intel's HLS flow; IV) Adding new auxiliary designs; V) Integrating published HLS data; VI) HLS tool version regression benchmarking. Code at https://github.com/sharc-lab/HLSFactory., Comment: Edit to "Section V.E" for proper attribution of open-source HLSyn, AutoDSE, and the Merlin compiler

Re-configurable Intelligent Surfaces (RIS) technology is increasingly becoming a potential component for next-generation wireless networks, offering enhanced performance in terms of throughput, spectral, and energy efficiency. However, the broadcast nature of RIS-assisted wireless communication makes it vulnerable to malicious attacks at the physical layer. At the same time, physical layer authentication is gaining popularity as a solution to secure wireless networks, thwarting different attacks such as cloning, spoofing, and impersonation by using the random features of the physical layer. In this paper, we investigate RIS-assisted wireless communication systems to unlock the potential of using RIS for physical layer authentication (PLA). In particular, we exploit two distinct features of the physical layer: pathloss and channel impulse response (CIR) for PLA in RIS-assisted wireless communication. We construct hypothesis tests for the estimated features and derive closed-form error expressions. Further, we consider the critical error, i.e., missed detection, as our objective function to minimize by optimizing the phase shift of the RIS pannel. We compare the performance of our proposed mechanisms with PLA schemes using the same features but with no RIS. Furthermore, we thoroughly evaluate our proposed schemes using performance metrics such as the probability of false alarm (PFA), the probability of missed detection (PMD), and the receiver operating characteristic (ROC) curves. The results demonstrate a clear positive impact of RIS on PLA, as it effectively reduces PMD values to zero when determining the optimal phase shift.

We report on a novel flow-based method for high-precision chemical detection that integrates quantum sensing with droplet microfluidics. We deploy nanodiamond particles hosting fluorescent nitrogen vacancy defects as quantum sensors in flowing, monodisperse, picoliter-volume microdroplets containing analyte molecules. ND motion within these microcompartments facilitates close sensor-analyte interaction and mitigates particle heterogeneity. Microdroplet flow rates are rapid (upto 4cm/s) and with minimal drift. Pairing this controlled flow with microwave control of NV electronic spins, we introduce a new noise-suppressed mode of Optically Detected Magnetic Resonance that is sensitive to chemical analytes while resilient against experimental variations, achieving detection of analyte-induced signals at an unprecedented level of a few hundredths of a percent of the ND fluorescence. We demonstrate its application to detecting paramagnetic ions in droplets with simultaneously low limit-of-detection and low analyte volumes, in a manner significantly better than existing technologies. This is combined with exceptional measurement stability over >103s and across hundreds of thousands of droplets, while utilizing minimal sensor volumes and incurring low ND costs (<$0.70 for an hour of operation). Additionally, we demonstrate using these droplets as micro-confinement chambers by co-encapsulating ND quantum sensors with analytes, including single cells. This versatility suggests wide-ranging applications, like single-cell metabolomics and real-time intracellular measurements in bioreactors. Our work paves the way for portable, high-sensitivity, amplification-free, chemical assays with high throughput; introduces a new chemical imaging tool for probing chemical reactions in microenvironments; and establishes the foundation for developing movable, arrayed quantum sensors through droplet microfluidics.

We introduce Spivavtor, a dataset, and instruction-tuned models for text editing focused on the Ukrainian language. Spivavtor is the Ukrainian-focused adaptation of the English-only CoEdIT model. Similar to CoEdIT, Spivavtor performs text editing tasks by following instructions in Ukrainian. This paper describes the details of the Spivavtor-Instruct dataset and Spivavtor models. We evaluate Spivavtor on a variety of text editing tasks in Ukrainian, such as Grammatical Error Correction (GEC), Text Simplification, Coherence, and Paraphrasing, and demonstrate its superior performance on all of them. We publicly release our best-performing models and data as resources to the community to advance further research in this space., Comment: Accepted to UNLP Workshop 2024

We introduce Harmonic Robustness, a powerful and intuitive method to test the robustness of any machine-learning model either during training or in black-box real-time inference monitoring without ground-truth labels. It is based on functional deviation from the harmonic mean value property, indicating instability and lack of explainability. We show implementation examples in low-dimensional trees and feedforward NNs, where the method reliably identifies overfitting, as well as in more complex high-dimensional models such as ResNet-50 and Vision Transformer where it efficiently measures adversarial vulnerability across image classes., Comment: 18 pages, 13 figures

Error Detection and Correction Codes (ECCs) are often used in digital designs to protect data integrity. Especially in safety-critical systems such as automotive electronics, ECCs are widely used and the verification of such complex logic becomes more critical considering the ISO 26262 safety standards. Exhaustive verification of ECC using formal methods has been a challenge given the high number of data bits to protect. As an example, for an ECC of 128 data bits with a possibility to detect up to four-bit errors, the combination of bit errors is given by 128C1 + 128C2 + 128C3 + 128C4 = 1.1 * 10^7. This vast analysis space often leads to bounded proof results. Moreover, the complexity and state-space increase further if the ECC has sequential encoding and decoding stages. To overcome such problems and sign-off the design with confidence within reasonable proof time, we present a pragmatic formal verification approach of complex ECC cores with several complexity reduction techniques and know-how that were learnt during the course of verification. We discuss using the linearity of the syndrome generator as a helper assertion, using the abstract model as glue logic to compare the RTL with the sequential version of the circuit, k-induction-based model checking and using mathematical relations captured as properties to simplify the verification in order to get an unbounded proof result within 24 hours of proof runtime., Comment: Published in DVCon U.S. 2023

Despite their remarkable successes, state-of-the-art language models face challenges in grasping certain important semantic details. This paper introduces the VISLA (Variance and Invariance to Semantic and Lexical Alterations) benchmark, designed to evaluate the semantic and lexical understanding of language models. VISLA presents a 3-way semantic (in)equivalence task with a triplet of sentences associated with an image, to evaluate both vision-language models (VLMs) and unimodal language models (ULMs). An evaluation involving 34 VLMs and 20 ULMs reveals surprising difficulties in distinguishing between lexical and semantic variations. Spatial semantics encoded by language models also appear to be highly sensitive to lexical information. Notably, text encoders of VLMs demonstrate greater sensitivity to semantic and lexical variations than unimodal text encoders. Our contributions include the unification of image-to-text and text-to-text retrieval tasks, an off-the-shelf evaluation without fine-tuning, and assessing LMs' semantic (in)variance in the presence of lexical alterations. The results highlight strengths and weaknesses across diverse vision and unimodal language models, contributing to a deeper understanding of their capabilities. % VISLA enables a rigorous evaluation, shedding light on language models' capabilities in handling semantic and lexical nuances. Data and code will be made available at https://github.com/Sri-Harsha/visla_benchmark.

This paper investigates the dynamics of crystalline clusters observed in Molecular Dynamics (MD) studies conducted earlier [Yadav, M., et al. Physical Review E, 107(5), 055214(2023)] for ultra-cold neutral plasmas. An external oscillatory forcing is applied for this purpose and the evolution is tracked with the help of MD simulations using the open source LAMMPS software. Interesting observations relating to cluster dynamics are presented. The formation of a pentagonal arrangement of particles is also reported., Comment: 9 pages, 12 figures

We introduce an efficient discretization of a novel fractional-order adaptive exponential (FrAdEx) integrate-and-fire model, which is used to study the fractional-order dynamics of neuronal activities. The discretization is based on extension of L1-type methods that can accurately handle the exponential growth and the spiking mechanism of the model. This new method is implicit and uses adaptive time stepping to robustly handle the stiff system that arises due to the exponential term. The implicit nonlinear system can be solved exactly, without the need for iterative methods, making the scheme efficient while maintaining accuracy. We present a complete error model for the numerical scheme that can be extended to other integrate-and-fire models with minor changes. To show the feasibility of our approach, the numerical method has been rigorously validated and used to investigate several different spiking oscillations of the model. We observed that the fractional-order model is capable of predicting biophysical activities, which are interpreted through phase diagrams describing the transition from one firing type to another. This simple model shows significant promise, as it has sufficient expressive dynamics to reproduce several features qualitatively from a biophysical dynamical perspective., Comment: 27 pages, 9 figures. v1: changed title, updated references

Modern System-on-Chip (SoC) designs are becoming more and more complex due to the technology upscaling. SoC designs often operate on multiple asynchronous clock domains, further adding to the complexity of the overall design. To make the devices power efficient, designers take a Globally-Asynchronous Locally-Synchronous (GALS) approach that creates multiple asynchronous domains. These Clock Domain Crossings (CDC) are prone to metastability effects, and functional verification of such CDC is very important to ensure that no bug escapes. Conventional verification methods, such as register transfer level (RTL) simulations and static timing analysis, are not enough to address these CDC issues, which may lead to verification gaps. Additionally, identifying these CDC-related bugs is very time-consuming and is one of the most common reasons for costly silicon re-spins. This paper is focused on the development of a pragmatic formal verification methodology to minimize the CDC issues by exercising Metastability Injection (MSI) in different CDC paths., Comment: Published in DVCon Europe 2023

Nowadays, a majority of System-on-Chips (SoCs) make use of Intellectual Property (IP) in order to shorten development cycles. When such IPs are developed, one of the main focuses lies in the high configurability of the design. This flexibility on the design side introduces the challenge of covering a huge state space of IP configurations on the verification side to ensure the functional correctness under every possible parameter setting. The vast number of possibilities does not allow a brute-force approach, and therefore, only a selected number of settings based on typical and extreme assumptions are usually verified. Especially in automotive applications, which need to follow the ISO 26262 functional safety standard, the requirement of covering all significant variants needs to be fulfilled in any case. State-of-the-Art existing verification techniques such as simulation-based verification and formal verification have challenges such as time-space explosion and state-space explosion respectively and therefore, lack behind in verifying highly configurable digital designs efficiently. This paper is focused on a semi-formal verification methodology for efficient configuration coverage of highly configurable digital designs. The methodology focuses on reduced runtime based on simulative and formal methods that allow high configuration coverage. The paper also presents the results when the developed methodology was applied on a highly configurable microprocessor IP and discusses the gained benefits., Comment: Published in DVCon U.S. 2021

As the demand for Internet of Things (IoT) and Human-to-Machine Interaction (HMI) increases, modern System-on-Chips (SoCs) offering such solutions are becoming increasingly complex. This intricate design poses significant challenges for verification, particularly when time-to-market is a crucial factor for consumer electronics products. This paper presents a case study based on our work to verify a complex Radio Detection And Ranging (RADAR) based SoC that performs on-chip sensing of human motion with millimetre accuracy. We leverage both formal and simulation-based methods to complement each other and achieve verification sign-off with high confidence. While employing a requirements-driven flow approach, we demonstrate the use of different verification methods to cater to multiple requirements and highlight our know-how from the project. Additionally, we used Machine Learning (ML) based methods, specifically the Xcelium ML tool from Cadence, to improve verification throughput., Comment: Published in DVCon Europe 2023

The rise of deep learning has marked significant progress in fields such as computer vision, natural language processing, and medical imaging, primarily through the adaptation of pre-trained models for specific tasks. Traditional fine-tuning methods, involving adjustments to all parameters, face challenges due to high computational and memory demands. This has led to the development of Parameter Efficient Fine-Tuning (PEFT) techniques, which selectively update parameters to balance computational efficiency with performance. This review examines PEFT approaches, offering a detailed comparison of various strategies highlighting applications across different domains, including text generation, medical imaging, protein modeling, and speech synthesis. By assessing the effectiveness of PEFT methods in reducing computational load, speeding up training, and lowering memory usage, this paper contributes to making deep learning more accessible and adaptable, facilitating its wider application and encouraging innovation in model optimization. Ultimately, the paper aims to contribute towards insights into PEFT's evolving landscape, guiding researchers and practitioners in overcoming the limitations of conventional fine-tuning approaches.

Tidal features provide signatures of recent mergers and offer a unique insight into the assembly history of galaxies. The Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST) will enable an unprecedentedly large survey of tidal features around millions of galaxies. To decipher the contributions of mergers to galaxy evolution it will be necessary to compare the observed tidal features with theoretical predictions. Therefore, we use cosmological hydrodynamical simulations NewHorizon, EAGLE, IllustrisTNG, and Magneticum to produce LSST-like mock images of $z\sim0$ galaxies ($z\sim0.2$ for NewHorizon) with $M_{\scriptstyle\star,\text{ 30 pkpc}}\geq10^{9.5}$ M$_{\scriptstyle\odot}$. We perform a visual classification to identify tidal features and classify their morphology. We find broadly good agreement between the simulations regarding their overall tidal feature fractions: $f_{\text{NewHorizon}}=0.40\pm0.06$, $f_{\text{EAGLE}}=0.37\pm0.01$, $f_{\text{TNG}}=0.32\pm0.01$ and $f_{\text{Magneticum}}=0.32\pm0.01$, and their specific tidal feature fractions. Furthermore, we find excellent agreement regarding the trends of tidal feature fraction with stellar and halo mass. All simulations agree in predicting that the majority of central galaxies of groups and clusters exhibit at least one tidal feature, while the satellite members rarely show such features. This agreement suggests that gravity is the primary driver of the occurrence of visually-identifiable tidal features in cosmological simulations, rather than subgrid physics or hydrodynamics. All predictions can be verified directly with LSST observations., Comment: 25 pages, 14 figures, accepted in MNRAS

FPGAs are a promising platform for accelerating Deep Learning (DL) applications, due to their high performance, low power consumption, and reconfigurability. Recently, the leading FPGA vendors have enhanced their architectures to more efficiently support the computational demands of DL workloads. However, the two most prominent AI-optimized FPGAs, i.e., AMD/Xilinx Versal ACAP and Intel Stratix 10 NX, employ significantly different architectural approaches. This paper presents novel systematic frameworks to optimize the performance of General Matrix Multiplication (GEMM), a fundamental operation in DL workloads, by exploiting the unique and distinct architectural characteristics of each FPGA. Our evaluation on GEMM workloads for int8 precision shows up to 77 and 68 TOPs (int8) throughput, with up to 0.94 and 1.35 TOPs/W energy efficiency for Versal VC1902 and Stratix 10 NX, respectively. This work provides insights and guidelines for optimizing GEMM-based applications on both platforms, while also delving into their programmability trade-offs and associated challenges., Comment: Accepted as full paper at FCCM 2024

BoLD is a new dispute resolution protocol that is designed to replace the originally deployed Arbitrum dispute resolution protocol. Unlike that protocol, BoLD is resistant to delay attacks. It achieves this resistance without a significant increase in onchain computation costs and with reduced staking costs.

Content moderation faces a challenging task as social media's ability to spread hate speech contrasts with its role in promoting global connectivity. With rapidly evolving slang and hate speech, the adaptability of conventional deep learning to the fluid landscape of online dialogue remains limited. In response, causality inspired disentanglement has shown promise by segregating platform specific peculiarities from universal hate indicators. However, its dependency on available ground truth target labels for discerning these nuances faces practical hurdles with the incessant evolution of platforms and the mutable nature of hate speech. Using confidence based reweighting and contrastive regularization, this study presents HATE WATCH, a novel framework of weakly supervised causal disentanglement that circumvents the need for explicit target labeling and effectively disentangles input features into invariant representations of hate. Empirical validation across platforms two with target labels and two without positions HATE WATCH as a novel method in cross platform hate speech detection with superior performance. HATE WATCH advances scalable content moderation techniques towards developing safer online communities.

This paper reviews the NTIRE 2024 challenge on image super-resolution ($\times$4), highlighting the solutions proposed and the outcomes obtained. The challenge involves generating corresponding high-resolution (HR) images, magnified by a factor of four, from low-resolution (LR) inputs using prior information. The LR images originate from bicubic downsampling degradation. The aim of the challenge is to obtain designs/solutions with the most advanced SR performance, with no constraints on computational resources (e.g., model size and FLOPs) or training data. The track of this challenge assesses performance with the PSNR metric on the DIV2K testing dataset. The competition attracted 199 registrants, with 20 teams submitting valid entries. This collective endeavour not only pushes the boundaries of performance in single-image SR but also offers a comprehensive overview of current trends in this field., Comment: NTIRE 2024 webpage: https://cvlai.net/ntire/2024. Code: https://github.com/zhengchen1999/NTIRE2024_ImageSR_x4

Quantifying bias in retrieval functions through document retrievability scores is vital for assessing recall-oriented retrieval systems. However, many studies investigating retrieval model bias lack validation of their query generation methods as accurate representations of retrievability for real users and their queries. This limitation results from the absence of established criteria for query generation in retrievability assessments. Typically, researchers resort to using frequent collocations from document corpora when no query log is available. In this study, we address the issue of reproducibility and seek to validate query generation methods by comparing retrievability scores generated from artificially generated queries to those derived from query logs. Our findings demonstrate a minimal or negligible correlation between retrievability scores from artificial queries and those from query logs. This suggests that artificially generated queries may not accurately reflect retrievability scores as derived from query logs. We further explore alternative query generation techniques, uncovering a variation that exhibits the highest correlation. This alternative approach holds promise for improving reproducibility when query logs are unavailable., Comment: Accepted at ECIR 2024

Deep learning (DL) is becoming the cornerstone of numerous applications both in datacenters and at the edge. Specialized hardware is often necessary to meet the performance requirements of state-of-the-art DL models, but the rapid pace of change in DL models and the wide variety of systems integrating DL make it impossible to create custom computer chips for all but the largest markets. Field-programmable gate arrays (FPGAs) present a unique blend of reprogrammability and direct hardware execution that make them suitable for accelerating DL inference. They offer the ability to customize processing pipelines and memory hierarchies to achieve lower latency and higher energy efficiency compared to general-purpose CPUs and GPUs, at a fraction of the development time and cost of custom chips. Their diverse high-speed IOs also enable directly interfacing the FPGA to the network and/or a variety of external sensors, making them suitable for both datacenter and edge use cases. As DL has become an ever more important workload, FPGA architectures are evolving to enable higher DL performance. In this article, we survey both academic and industrial FPGA architecture enhancements for DL. First, we give a brief introduction on the basics of FPGA architecture and how its components lead to strengths and weaknesses for DL applications. Next, we discuss different styles of DL inference accelerators on FPGA, ranging from model-specific dataflow styles to software-programmable overlay styles. We survey DL-specific enhancements to traditional FPGA building blocks such as logic blocks, arithmetic circuitry, and on-chip memories, as well as new in-fabric DL-specialized blocks for accelerating tensor computations. Finally, we discuss hybrid devices that combine processors and coarse-grained accelerator blocks with FPGA-like interconnect and networks-on-chip, and highlight promising future research directions.

This paper investigates the potential of blockchain-based fan tokens, a class of crypto asset that grants holders access to voting on club decisions and other perks, as a mechanism for stimulating democratized decision-making and fan engagement in the sports and esports sectors. By utilizing an extensive dataset of 3,576 fan token polls, we reveal that fan tokens engage an average of 4,003 participants per poll, representing around 50% of token holders, underscoring their relative effectiveness in boosting fan engagement. The analyses identify significant determinants of fan token poll participation, including levels of voter (dis-)agreement, poll type, sports sectors, demographics, and club-level factors. This study provides valuable stakeholder insights into the current state of adoption and voting trends for fan token polls. It also suggests strategies for increasing fan engagement, thereby optimizing the utility of fan tokens in sports. Moreover, we highlight the broader applicability of fan token principles to any community, brand, or organization focused on customer engagement, suggesting a wider potential for this digital innovation., Comment: 45 pages, 8 figures, 4 tables, 1 appendix figure, 1 appendix table

We propose the first-ever complete, model-agnostic search strategy based on the optimal anomaly score, for new physics on the tails of distributions. Signal sensitivity is achieved via a classifier trained on auxiliary features in a weakly-supervised fashion, and backgrounds are predicted using the ABCD method in the classifier output and the primary tail feature. The independence between the classifier output and the tail feature required for ABCD is achieved by first training a conditional normalizing flow that yields a decorrelated version of the auxiliary features; the classifier is then trained on these features. Both the signal sensitivity and background prediction require a sample of events accurately approximating the SM background; we assume this can be furnished by closely related control processes in the data or by accurate simulations, as is the case in countless conventional analyses. The viability of our approach is demonstrated for signatures consisting of (mono)jets and missing transverse energy, where the main SM background is $Z(\nu \nu) +\text{jets}$, and the data-driven control process is $\gamma+\text{jets}$., Comment: 9 pages, 9 figures

Estimating the out-of-distribution performance in regimes where labels are scarce is critical to safely deploy foundation models. Recently, it was shown that ensembles of neural networks observe the phenomena "agreement-on-the-line", which can be leveraged to reliably predict OOD performance without labels. However, in contrast to classical neural networks that are trained on in-distribution data from scratch for numerous epochs, foundation models undergo minimal finetuning from heavily pretrained weights, which may reduce the ensemble diversity needed to observe agreement-on-the-line. In our work, we demonstrate that when lightly finetuning multiple runs from a single foundation model, the choice of randomness during training (linear head initialization, data ordering, and data subsetting) can lead to drastically different levels of agreement-on-the-line in the resulting ensemble. Surprisingly, only random head initialization is able to reliably induce agreement-on-the-line in finetuned foundation models across vision and language benchmarks. Second, we demonstrate that ensembles of multiple foundation models pretrained on different datasets but finetuned on the same task can also show agreement-on-the-line. In total, by careful construction of a diverse ensemble, we can utilize agreement-on-the-line-based methods to predict the OOD performance of foundation models with high precision.

The honeycomb Kitaev model describes a $Z_2$ spin liquid with topological order and fractionalized excitations consisting of gapped $\pi$-fluxes and free Majorana fermions. Competing interactions, even when not very strong, are known to destabilize the Kitaev spin liquid. Magnetic fields are a convenient parameter for tuning between different phases of the Kitaev systems, and have even been investigated for potentially counteracting the effects of other destabilizing interactions leading to a revival of the topological phase. Here we review the progress in understanding the effects of magnetic fields on some of the perturbed Kitaev systems, particularly on fractionalization and topological order., Comment: 24 pages, 2 figures, For JPCM topical reviews: comments are welcome

The widespread adoption of Large Language Models (LLMs) has facilitated numerous benefits. However, hallucination is a significant concern. In response, Retrieval Augmented Generation (RAG) has emerged as a highly promising paradigm to improve LLM outputs by grounding them in factual information. RAG relies on textual entailment (TE) or similar methods to check if the text produced by LLMs is supported or contradicted, compared to retrieved documents. This paper argues that conventional TE methods are inadequate for spotting hallucinations in content generated by LLMs. For instance, consider a prompt about the 'USA's stance on the Ukraine war''. The AI-generated text states, ...U.S. President Barack Obama says the U.S. will not put troops in Ukraine...'' However, during the war the U.S. president is Joe Biden which contradicts factual reality. Moreover, current TE systems are unable to accurately annotate the given text and identify the exact portion that is contradicted. To address this, we introduces a new type of TE called ``Factual Entailment (FE).'', aims to detect factual inaccuracies in content generated by LLMs while also highlighting the specific text segment that contradicts reality. We present FACTOID (FACTual enTAILment for hallucInation Detection), a benchmark dataset for FE. We propose a multi-task learning (MTL) framework for FE, incorporating state-of-the-art (SoTA) long text embeddings such as e5-mistral-7b-instruct, along with GPT-3, SpanBERT, and RoFormer. The proposed MTL architecture for FE achieves an avg. 40\% improvement in accuracy on the FACTOID benchmark compared to SoTA TE methods. As FE automatically detects hallucinations, we assessed 15 modern LLMs and ranked them using our proposed Auto Hallucination Vulnerability Index (HVI_auto). This index quantifies and offers a comparative scale to evaluate and rank LLMs according to their hallucinations.

Hallucination has emerged as the most vulnerable aspect of contemporary Large Language Models (LLMs). In this paper, we introduce the Sorry, Come Again (SCA) prompting, aimed to avoid LLM hallucinations by enhancing comprehension through: (i) optimal paraphrasing and (ii) injecting [PAUSE] tokens to delay LLM generation. First, we provide an in-depth analysis of linguistic nuances: formality, readability, and concreteness of prompts for 21 LLMs, and elucidate how these nuances contribute to hallucinated generation. Prompts with lower readability, formality, or concreteness pose comprehension challenges for LLMs, similar to those faced by humans. In such scenarios, an LLM tends to speculate and generate content based on its imagination (associative memory) to fill these information gaps. Although these speculations may occasionally align with factual information, their accuracy is not assured, often resulting in hallucination. Recent studies reveal that an LLM often neglects the middle sections of extended prompts, a phenomenon termed as lost in the middle. While a specific paraphrase may suit one LLM, the same paraphrased version may elicit a different response from another LLM. Therefore, we propose an optimal paraphrasing technique to identify the most comprehensible paraphrase of a given prompt, evaluated using Integrated Gradient (and its variations) to guarantee that the LLM accurately processes all words. While reading lengthy sentences, humans often pause at various points to better comprehend the meaning read thus far. We have fine-tuned an LLM with injected [PAUSE] tokens, allowing the LLM to pause while reading lengthier prompts. This has brought several key contributions: (i) determining the optimal position to inject [PAUSE], (ii) determining the number of [PAUSE] tokens to be inserted, and (iii) introducing reverse proxy tuning to fine-tune the LLM for [PAUSE] insertion.

In this brief paper, the authors study the tuning curves of starburst amacrine cells (SACs) and introduce a quantity called the irresolution or ambiguity of a SAC. They show that the rate of data generated by a starburst amacrine cell is inversely proportional to its irresolution. This is done by providing bounds on the rate required to encode the generated data. This technique can be applied to different cell types with different tuning curves. In this manner, information theoretic views can be introduced to the cases of biological cells which are not normally considered as transmitters of information as say retinal ganglion cells are. The intuition that even such cells generate information is thus quantified., Comment: 6 pages, 7 figures

Over the past few years, Text-to-Image (T2I) generation approaches based on diffusion models have gained significant attention. However, vanilla diffusion models often suffer from spelling inaccuracies in the text displayed within the generated images. The capability to generate visual text is crucial, offering both academic interest and a wide range of practical applications. To produce accurate visual text images, state-of-the-art techniques adopt a glyph-controlled image generation approach, consisting of a text layout generator followed by an image generator that is conditioned on the generated text layout. Nevertheless, our study reveals that these models still face three primary challenges, prompting us to develop a testbed to facilitate future research. We introduce a benchmark, LenCom-Eval, specifically designed for testing models' capability in generating images with Lengthy and Complex visual text. Subsequently, we introduce a training-free framework to enhance the two-stage generation approaches. We examine the effectiveness of our approach on both LenCom-Eval and MARIO-Eval benchmarks and demonstrate notable improvements across a range of evaluation metrics, including CLIPScore, OCR precision, recall, F1 score, accuracy, and edit distance scores. For instance, our proposed framework improves the backbone model, TextDiffuser, by more than 23\% and 13.5\% in terms of OCR word F1 on LenCom-Eval and MARIO-Eval, respectively. Our work makes a unique contribution to the field by focusing on generating images with long and rare text sequences, a niche previously unexplored by existing literature, Comment: Accepted at WACV 2025

Modern hardware designs have grown increasingly efficient and complex. However, they are often susceptible to Common Weakness Enumerations (CWEs). This paper is focused on the formal verification of CWEs in a dataset of hardware designs written in SystemVerilog from Regenerative Artificial Intelligence (AI) powered by Large Language Models (LLMs). We applied formal verification to categorize each hardware design as vulnerable or CWE-free. This dataset was generated by 4 different LLMs and features a unique set of designs for each of the 10 CWEs we target in our paper. We have associated the identified vulnerabilities with CWE numbers for a dataset of 60,000 generated SystemVerilog Register Transfer Level (RTL) code. It was also found that most LLMs are not aware of any hardware CWEs; hence they are usually not considered when generating the hardware code. Our study reveals that approximately 60% of the hardware designs generated by LLMs are prone to CWEs, posing potential safety and security risks. The dataset could be ideal for training LLMs and Machine Learning (ML) algorithms to abstain from generating CWE-prone hardware designs., Comment: Published in DVCon U.S. 2024

National football teams increasingly issue tradeable blockchain-based fan tokens to strategically enhance fan engagement. This study investigates the impact of 2022 World Cup matches on the dynamic performance of each team's fan token. The event study uncovers fan token returns surged six months before the World Cup, driven by positive anticipation effects. However, intraday analysis reveals a reversal of fan token returns consistently declining and trading volumes rising as matches unfold. To explain findings, we uncover asymmetries whereby defeats in high-stake matches caused a plunge in fan token returns, compared to low-stake matches, intensifying in magnitude for knockout matches. Contrarily, victories enhance trading volumes, reflecting increased market activity without a corresponding positive effect on returns. We align findings with the classic market adage "buy the rumor, sell the news," unveiling cognitive biases and nuances in investor sentiment, cautioning the dichotomy of pre-event optimism and subsequent performance declines., Comment: 31 pages, 5 tables, 1 figure

We present the discovery of a unique Lyman-continuum (LyC) emitter at z=3.088. The LyC emission were detected using the Hubble Space Telescope (HST) WFC3/UVIS F336W filter, covering a rest-frame wavelength range of 760-900 Angstrom. The peak signal-to-noise ratio (SNR) of LyC emission is 3.9 in a r=0.24'' aperture and is spatially offset by 0.29''+/-0.04'' (~ 2.2+/-0.3 kpc) from the rest-UV emission peak (F606W). By combining imaging and spectroscopic data from the James Webb Space Telescope (JWST) JADES, FRESCO and JEMS surveys, along with VLT/MUSE data from the MXDF survey, we estimate that the probability of random alignment with an interloper galaxy causing the LyC emission is less than 6x10^-5. The interstellar medium (ISM) conditions in the galaxy are similar to other LyC emitters at high redshift (12+log(O/H)=7.79+/-0.06, logU =-3.27+/-0.14, O32 = 3.65+/-0.22), although the single-peaked Lyman-alpha profile and lack of rest-UV emission lines suggest an optically thick ISM. We think that LyC photons are leaking through a narrow cone of optically thin neutral ISM, most likely created by a past merger (as evidenced by medium-band F210M and F182M images). Using the escape fraction constraints from individual leakers and a simple model, we estimate that the opening half-angle of ionization cones can be as low as 16^deg (2% ionised fraction) to reproduce some of the theoretical constraints on the average escape fraction for galaxies. The narrow opening angle required can explain the low number density of confirmed LyC leakers., Comment: 12 pages, 8 figures, and 1 table. Submitted to The Astrophysical Journal

When interacting with Retrieval-Augmented Generation (RAG)-based conversational agents, the users must carefully craft their queries to be understood correctly. Yet, understanding the system's capabilities can be challenging for the users, leading to ambiguous questions that necessitate further clarification. This work aims to bridge the gap by developing a suggestion question generator. To generate suggestion questions, our approach involves utilizing dynamic context, which includes both dynamic few-shot examples and dynamically retrieved contexts. Through experiments, we show that the dynamic contexts approach can generate better suggestion questions as compared to other prompting approaches., Comment: 4 pages , 1 figure, accepted at WWW'24 workshop PromptEng

Moire pattern frequently appears in photographs captured with mobile devices and digital cameras, potentially degrading image quality. Despite recent advancements in computer vision, image demoire'ing remains a challenging task due to the dynamic textures and variations in colour, shape, and frequency of moire patterns. Most existing methods struggle to generalize to unseen datasets, limiting their effectiveness in removing moire patterns from real-world scenarios. In this paper, we propose a novel lightweight architecture, AADNet (Attention Aware Demoireing Network), for high-resolution image demoire'ing that effectively works across different frequency bands and generalizes well to unseen datasets. Extensive experiments conducted on the UHDM dataset validate the effectiveness of our approach, resulting in high-fidelity images., Comment: Due to unauthorized access and upload, this paper has been withdrawn. It does not reflect the contributions or approval

Generative Adversarial Networks (GANs) have proven to exhibit remarkable performance and are widely used across many generative computer vision applications. However, the unprecedented demand for the deployment of GANs on resource-constrained edge devices still poses a challenge due to huge number of parameters involved in the generation process. This has led to focused attention on the area of compressing GANs. Most of the existing works use knowledge distillation with the overhead of teacher dependency. Moreover, there is no ability to control the degree of compression in these methods. Hence, we propose CoroNet-GAN for compressing GAN using the combined strength of differentiable pruning method via hypernetworks. The proposed method provides the advantage of performing controllable compression while training along with reducing training time by a substantial factor. Experiments have been done on various conditional GAN architectures (Pix2Pix and CycleGAN) to signify the effectiveness of our approach on multiple benchmark datasets such as Edges-to-Shoes, Horse-to-Zebra and Summer-to-Winter. The results obtained illustrate that our approach succeeds to outperform the baselines on Zebra-to-Horse and Summer-to-Winter achieving the best FID score of 32.3 and 72.3 respectively, yielding high-fidelity images across all the datasets. Additionally, our approach also outperforms the state-of-the-art methods in achieving better inference time on various smart-phone chipsets and data-types making it a feasible solution for deployment on edge devices.

In the midst of widespread misinformation and disinformation through social media and the proliferation of AI-generated texts, it has become increasingly difficult for people to validate and trust information they encounter. Many fact-checking approaches and tools have been developed, but they often lack appropriate explainability or granularity to be useful in various contexts. A text validation method that is easy to use, accessible, and can perform fine-grained evidence attribution has become crucial. More importantly, building user trust in such a method requires presenting the rationale behind each prediction, as research shows this significantly influences people's belief in automated systems. Localizing and bringing users' attention to the specific problematic content is also paramount, instead of providing simple blanket labels. In this paper, we present ClaimVer, a human-centric framework tailored to meet users' informational and verification needs by generating rich annotations and thereby reducing cognitive load. Designed to deliver comprehensive evaluations of texts, it highlights each claim, verifies it against a trusted knowledge graph (KG), presents the evidence, and provides succinct, clear explanations for each claim prediction. Finally, our framework introduces an attribution score, enhancing applicability across a wide range of downstream tasks., Comment: EMNLP 2024 Findings

The widespread presence of hateful languages on social media has resulted in adverse effects on societal well-being. As a result, addressing this issue with high priority has become very important. Hate speech or offensive languages exist in both explicit and implicit forms, with the latter being more challenging to detect. Current research in this domain encounters several challenges. Firstly, the existing datasets primarily rely on the collection of texts containing explicit offensive keywords, making it challenging to capture implicitly offensive contents that are devoid of these keywords. Secondly, common methodologies tend to focus solely on textual analysis, neglecting the valuable insights that community information can provide. In this research paper, we introduce a novel dataset OffensiveLang, a community based implicit offensive language dataset generated by ChatGPT 3.5 containing data for 38 different target groups. Despite limitations in generating offensive texts using ChatGPT due to ethical constraints, we present a prompt-based approach that effectively generates implicit offensive languages. To ensure data quality, we evaluate the dataset with human. Additionally, we employ a prompt-based zero-shot method with ChatGPT and compare the detection results between human annotation and ChatGPT annotation. We utilize existing state-of-the-art models to see how effective they are in detecting such languages. The dataset is available here: https://github.com/AmitDasRup123/OffensiveLang

The use of LLMs in natural language reasoning has shown mixed results, sometimes rivaling or even surpassing human performance in simpler classification tasks while struggling with social-cognitive reasoning, a domain where humans naturally excel. These differences have been attributed to many factors, such as variations in prompting and the specific LLMs used. However, no reasons appear conclusive, and no clear mechanisms have been established in prior work. In this study, we empirically evaluate how role-playing prompting influences Theory-of-Mind (ToM) reasoning capabilities. Grounding our rsearch in psychological theory, we propose the mechanism that, beyond the inherent variance in the complexity of reasoning tasks, performance differences arise because of socially-motivated prompting differences. In an era where prompt engineering with role-play is a typical approach to adapt LLMs to new contexts, our research advocates caution as models that adopt specific personas might potentially result in errors in social-cognitive reasoning., Comment: 8 pages

Large Language Models (LLMs) have become a cornerstone in the field of Natural Language Processing (NLP), offering transformative capabilities in understanding and generating human-like text. However, with their rising prominence, the security and vulnerability aspects of these models have garnered significant attention. This paper presents a comprehensive survey of the various forms of attacks targeting LLMs, discussing the nature and mechanisms of these attacks, their potential impacts, and current defense strategies. We delve into topics such as adversarial attacks that aim to manipulate model outputs, data poisoning that affects model training, and privacy concerns related to training data exploitation. The paper also explores the effectiveness of different attack methodologies, the resilience of LLMs against these attacks, and the implications for model integrity and user trust. By examining the latest research, we provide insights into the current landscape of LLM vulnerabilities and defense mechanisms. Our objective is to offer a nuanced understanding of LLM attacks, foster awareness within the AI community, and inspire robust solutions to mitigate these risks in future developments.

This is a summary paper of a use case of a Robotdog dedicated to guide visually impaired people in complex environment like a smart intersection. In such scenarios, the Robotdog has to autonomously decide whether it is safe to cross the intersection or not in order to further guide the human. We leverage data sharing and collaboration between the Robotdog and other autonomous systems operating in the same environment. We propose a system architecture for autonomous systems through a separation of a collaborative decision layer, to enable collective decision making processes, where data about the environment, relevant to the Robotdog decision, together with evidences for trustworthiness about other systems and the environment are shared.

We propose a generic multi-robot planning mechanism that combines an optimal task planner and an optimal path planner to provide a scalable solution for complex multi-robot planning problems. The Integrated planner, through the interaction of the task planner and the path planner, produces optimal collision-free trajectories for the robots. We illustrate our general algorithm on an object pick-and-drop planning problem in a warehouse scenario where a group of robots is entrusted with moving objects from one location to another in the workspace. We solve the task planning problem by reducing it into an SMT-solving problem and employing the highly advanced SMT solver Z3 to solve it. To generate collision-free movement of the robots, we extend the state-of-the-art algorithm Conflict Based Search with Precedence Constraints with several domain-specific constraints. We evaluate our integrated task and path planner extensively on various instances of the object pick-and-drop planning problem and compare its performance with a state-of-the-art multi-robot classical planner. Experimental results demonstrate that our planning mechanism can deal with complex planning problems and outperforms a state-of-the-art classical planner both in terms of computation time and the quality of the generated plan.

We have witnessed lately a rapid proliferation of advanced Large Language Models (LLMs) capable of generating high-quality text. While these LLMs have revolutionized text generation across various domains, they also pose significant risks to the information ecosystem, such as the potential for generating convincing propaganda, misinformation, and disinformation at scale. This paper offers a review of AI-generated text forensic systems, an emerging field addressing the challenges of LLM misuses. We present an overview of the existing efforts in AI-generated text forensics by introducing a detailed taxonomy, focusing on three primary pillars: detection, attribution, and characterization. These pillars enable a practical understanding of AI-generated text, from identifying AI-generated content (detection), determining the specific AI model involved (attribution), and grouping the underlying intents of the text (characterization). Furthermore, we explore available resources for AI-generated text forensics research and discuss the evolving challenges and future directions of forensic systems in an AI era.